{ "title": "AKI Superintelligence Index v1.2 — Top 100 AI Models Ledger", "version": "1.2.0", "governance": "ZIP-1.0", "generated_at": "2026-10-09", "canonical_url": "https://aki1k.com/superintelligence-index", "api_endpoint": "https://api.aki1k.com/v1/superintelligence", "total_models": 100, "benchmark_feeds_count": 23, "benchmark_feeds": [ { "id": "gdpval", "name": "Artificial Analysis GDPval", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Economically-weighted frontier intelligence index.", "source_authority": "Artificial Analysis" }, { "id": "epoch_eci", "name": "Epoch AI ECI", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Empirical capability index tracking historical growth rate.", "source_authority": "Epoch AI" }, { "id": "gpqa_diamond", "name": "GPQA Diamond", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Graduate-level Google-proof PhD STEM questions.", "source_authority": "NYU / Anthropic" }, { "id": "hle", "name": "Humanity's Last Exam", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Closed-world multi-disciplinary frontier evaluation.", "source_authority": "CAIS & Scale AI" }, { "id": "frontier_math", "name": "FrontierMath", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Unsolved research-grade mathematics challenges.", "source_authority": "Epoch AI" }, { "id": "arc_agi_3", "name": "ARC-AGI-3", "category": "Intelligence", "dimension_weight": 0.35, "single_source_cap": 0.12, "description": "Fluid visual-logic generalization beyond training distribution.", "source_authority": "François Chollet" }, { "id": "swe_bench_verified", "name": "SWE-bench Verified/Pro", "category": "Coding", "dimension_weight": 0.2, "single_source_cap": 0.12, "description": "Real-world GitHub issue resolution with unit tests.", "source_authority": "Princeton NLP" }, { "id": "terminal_bench", "name": "Terminal-Bench 4.0", "category": "Coding", "dimension_weight": 0.2, "single_source_cap": 0.12, "description": "Autonomous bash environment and CLI debugging loop.", "source_authority": "SWE & Agentic Evaluators" }, { "id": "live_code_bench", "name": "LiveCodeBench", "category": "Coding", "dimension_weight": 0.2, "single_source_cap": 0.12, "description": "Contamination-free competitive programming problems.", "source_authority": "UC Berkeley" }, { "id": "scicode", "name": "SciCode", "category": "Coding", "dimension_weight": 0.2, "single_source_cap": 0.12, "description": "Multi-step computational scientific problem solving.", "source_authority": "Stanford / MIT" }, { "id": "osworld", "name": "OSWorld 2.0", "category": "Agentic", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Full desktop OS GUI multimodal autonomy benchmark.", "source_authority": "CMU / Stanford" }, { "id": "browse_comp", "name": "BrowseComp", "category": "Agentic", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Multi-tab live web browsing and enterprise navigation.", "source_authority": "Browser Agent Guild" }, { "id": "automation_bench", "name": "AutomationBench", "category": "Agentic", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Enterprise workflow API and tool chaining autonomy.", "source_authority": "Automation Alliance" }, { "id": "gaia", "name": "GAIA General Assistants", "category": "Agentic", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Complex multi-modal multimodal assistant tasks.", "source_authority": "Meta AI / Hugging Face" }, { "id": "vectara_hallucination", "name": "Vectara Hallucination Index", "category": "Reliability", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Empirical factual error and hallucination rate.", "source_authority": "Vectara" }, { "id": "facts_grounding", "name": "FACTS Grounding Index", "category": "Reliability", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Retrieval-augmented grounding verification rate.", "source_authority": "Grounding Council" }, { "id": "uptime_monitor", "name": "Enterprise Uptime Monitor", "category": "Reliability", "dimension_weight": 0.15, "single_source_cap": 0.12, "description": "Live SLA availability and 5xx error rate audit.", "source_authority": "Global Edge Health" }, { "id": "openrouter_pricing", "name": "OpenRouter Token Pricing ($/M)", "category": "Efficiency", "dimension_weight": 0.1, "single_source_cap": 0.12, "description": "Blended input/output economic cost efficiency.", "source_authority": "OpenRouter API" }, { "id": "aa_throughput", "name": "Artificial Analysis Throughput", "category": "Efficiency", "dimension_weight": 0.1, "single_source_cap": 0.12, "description": "Output tokens per second generation velocity.", "source_authority": "Artificial Analysis" }, { "id": "hf_license", "name": "Hugging Face License Audit", "category": "Openness", "dimension_weight": 0.05, "single_source_cap": 0.12, "description": "Commercial usage freedom and weights inspectability.", "source_authority": "Hugging Face Hub" }, { "id": "epoch_reproducibility", "name": "Epoch Reproducibility Audit", "category": "Openness", "dimension_weight": 0.05, "single_source_cap": 0.12, "description": "Third-party open weights reproducibility audit.", "source_authority": "Epoch AI" }, { "id": "lmsys_arena", "name": "LMSYS Chatbot Arena", "category": "Signal Meta", "dimension_weight": 0, "single_source_cap": 0.12, "description": "Crowdsourced pairwise Elo (discounted 70%; 30% signal).", "source_authority": "LMSYS Org" }, { "id": "openrouter_volume", "name": "OpenRouter Volume Ranking", "category": "Signal Meta", "dimension_weight": 0, "single_source_cap": 0.12, "description": "Actual developer token demand volume (0% quality signal).", "source_authority": "OpenRouter Telemetry" } ], "safeguards": { "single_source_cap": "12%", "vendor_self_reported_cap": "50%", "freeze_window": "48h", "freshness_multiplier": "1.00x - 1.05x", "hype_divergence_penalty": "-0.3 to -1.6 pts" }, "models": [ { "rank": 1, "id": "gpt-6-astra-max", "name": "GPT-6 Astra Max", "provider": "OpenAI", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 92.4, "aa_intelligence": 52.7, "confidence_interval_95": 0.3, "pricing_input_per_m": 10, "pricing_output_per_m": 50, "openness": "Proprietary", "key_strength": "FrontierMath, HLE Diamond (60.6%) & ExploitBench", "value_index": 42.1, "context_window": "4096K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-09-18", "benchmarks": { "gdpval": 53.2, "epoch_eci": 52.8, "gpqa_diamond": 88.4, "hle": 60.6, "frontier_math": 34.2, "arc_agi_3": 82.5, "swe_bench_verified": 84.6, "terminal_bench": 91.2, "live_code_bench": 85, "scicode": 79.4, "osworld": 74.2, "browse_comp": 82.6, "automation_bench": 86.4, "gaia": 81.2, "vectara_hallucination": 1.4, "facts_grounding": 97.4, "uptime_pct": 99.98, "pricing_input_per_m": 10, "pricing_output_per_m": 50, "tokens_per_sec": 65.4, "hf_license_audit": 10, "epoch_reproducibility": 35, "lmsys_arena_elo": 1412, "openrouter_rank": 1 }, "replication_count": 6, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://openai.com/index/gpt-6-astra-frontier-evaluations/", "official_docs_url": "https://platform.openai.com/docs/models/gpt-6-astra", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-6-astra-max", "sha256": "7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b" }, { "rank": 2, "id": "gemini-4-argon-high", "name": "Gemini 4 Argon High", "provider": "Google DeepMind", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 91.7, "aa_intelligence": 52.6, "confidence_interval_95": 0.4, "pricing_input_per_m": 3.5, "pricing_output_per_m": 12, "openness": "Proprietary", "key_strength": "Multimodal Sparse MoE · DeepSWE 77.9%", "value_index": 78.4, "context_window": "2000K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-08-12", "benchmarks": { "gdpval": 53, "epoch_eci": 52.5, "gpqa_diamond": 86.9, "hle": 58.2, "frontier_math": 31.8, "arc_agi_3": 84.1, "swe_bench_verified": 82.4, "terminal_bench": 88.5, "live_code_bench": 83.2, "scicode": 77.9, "osworld": 76.8, "browse_comp": 84.1, "automation_bench": 84.5, "gaia": 80.5, "vectara_hallucination": 1.5, "facts_grounding": 97.1, "uptime_pct": 99.99, "pricing_input_per_m": 3.5, "pricing_output_per_m": 12, "tokens_per_sec": 92, "hf_license_audit": 10, "epoch_reproducibility": 35, "lmsys_arena_elo": 1406, "openrouter_rank": 3 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://deepmind.google/technologies/gemini/argon/", "official_docs_url": "https://ai.google.dev/gemini-api/docs/models/gemini-4-argon", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-4-argon-high", "sha256": "6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c" }, { "rank": 3, "id": "claude-opus-5-5-max", "name": "Claude Opus 5.5 Max", "provider": "Anthropic", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 90.2, "aa_intelligence": 57.6, "confidence_interval_95": 0.5, "pricing_input_per_m": 4, "pricing_output_per_m": 20, "openness": "Proprietary", "key_strength": "SWE-Bench Pro V2 (99.4%) · Agentic Coherence", "value_index": 38.2, "context_window": "1000K", "delta_prev": "-0.5", "trend": "down", "release_date": "2026-07-29", "benchmarks": { "gdpval": 58.1, "epoch_eci": 57.2, "gpqa_diamond": 89.2, "hle": 54.8, "frontier_math": 29.5, "arc_agi_3": 79.8, "swe_bench_verified": 88.9, "terminal_bench": 94.6, "live_code_bench": 88.4, "scicode": 82.6, "osworld": 71.5, "browse_comp": 80.2, "automation_bench": 88.2, "gaia": 83.4, "vectara_hallucination": 1.6, "facts_grounding": 96.9, "uptime_pct": 99.95, "pricing_input_per_m": 4, "pricing_output_per_m": 20, "tokens_per_sec": 72.8, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1410, "openrouter_rank": 2 }, "replication_count": 6, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://www.anthropic.com/news/claude-5-5-family", "official_docs_url": "https://docs.anthropic.com/claude/reference/opus-5-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-opus-5-5-max", "sha256": "5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b" }, { "rank": 4, "id": "kimi-k3-max", "name": "Kimi K3 Max", "provider": "Moonshot AI", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 89.9, "aa_intelligence": 43.6, "confidence_interval_95": 0.6, "pricing_input_per_m": 2.2, "pricing_output_per_m": 8.8, "openness": "Proprietary", "key_strength": "1M Retrieval Precision & Long Thinking Loop", "value_index": 71.3, "context_window": "1000K", "delta_prev": "+0.2", "trend": "flat", "release_date": "2026-06-15", "benchmarks": { "gdpval": 44.2, "epoch_eci": 43.8, "gpqa_diamond": 83.4, "hle": 48.6, "frontier_math": 26.4, "arc_agi_3": 76.5, "swe_bench_verified": 76.2, "terminal_bench": 82.4, "live_code_bench": 78.6, "scicode": 74.2, "osworld": 68.4, "browse_comp": 78.5, "automation_bench": 81.2, "gaia": 76.8, "vectara_hallucination": 2.1, "facts_grounding": 95.8, "uptime_pct": 99.9, "pricing_input_per_m": 2.2, "pricing_output_per_m": 8.8, "tokens_per_sec": 84.2, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1378, "openrouter_rank": 7 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://kimi.moonshot.cn/k3", "official_docs_url": "https://platform.moonshot.cn/docs", "eval_audit_url": "https://artificialanalysis.ai/models/kimi-k3-max", "sha256": "4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c" }, { "rank": 5, "id": "qwen-3-8-max", "name": "Qwen 3.8 Max", "provider": "Alibaba Cloud", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 89.1, "aa_intelligence": 53.4, "confidence_interval_95": 0.5, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.2, "openness": "Open Weights", "key_strength": "2.4T MoE · Multilingual Agentic Autonomy", "value_index": 82.7, "context_window": "1000K", "delta_prev": "+0.8", "trend": "up", "release_date": "2026-08-01", "benchmarks": { "gdpval": 54, "epoch_eci": 53.2, "gpqa_diamond": 85.1, "hle": 51.4, "frontier_math": 27.8, "arc_agi_3": 78.2, "swe_bench_verified": 79.4, "terminal_bench": 84.8, "live_code_bench": 80.2, "scicode": 75.8, "osworld": 70.2, "browse_comp": 79.4, "automation_bench": 82.8, "gaia": 78.4, "vectara_hallucination": 2, "facts_grounding": 96.2, "uptime_pct": 99.92, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.2, "tokens_per_sec": 88.5, "hf_license_audit": 85, "epoch_reproducibility": 75, "lmsys_arena_elo": 1395, "openrouter_rank": 5 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://qwenlm.github.io/blog/qwen3.8-max/", "official_docs_url": "https://help.aliyun.com/document_detail/qwen3-8", "eval_audit_url": "https://artificialanalysis.ai/models/qwen-3-8-max", "sha256": "3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d" }, { "rank": 6, "id": "gpt-6-1-sol-max", "name": "GPT-6.1 Sol Max", "provider": "OpenAI", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 88.9, "aa_intelligence": 51.8, "confidence_interval_95": 0.4, "pricing_input_per_m": 2, "pricing_output_per_m": 10, "openness": "Proprietary", "key_strength": "Distilled Frontier + Speculative Decoding", "value_index": 75.6, "context_window": "1100K", "delta_prev": "+0.6", "trend": "up", "release_date": "2026-08-25", "benchmarks": { "gdpval": 52.4, "epoch_eci": 51.9, "gpqa_diamond": 84.8, "hle": 49.8, "frontier_math": 26.9, "arc_agi_3": 77.4, "swe_bench_verified": 78.2, "terminal_bench": 83.5, "live_code_bench": 79.4, "scicode": 74.8, "osworld": 69.4, "browse_comp": 78.2, "automation_bench": 81.6, "gaia": 77.2, "vectara_hallucination": 1.8, "facts_grounding": 96.5, "uptime_pct": 99.96, "pricing_input_per_m": 2, "pricing_output_per_m": 10, "tokens_per_sec": 124, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1388, "openrouter_rank": 6 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://openai.com/index/gpt-6-1-sol/", "official_docs_url": "https://platform.openai.com/docs/models/gpt-6-1-sol", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-6-1-sol-max", "sha256": "2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e" }, { "rank": 7, "id": "claude-sonnet-5-5-max", "name": "Claude Sonnet 5.5 Max", "provider": "Anthropic", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 88.8, "aa_intelligence": 56, "confidence_interval_95": 0.4, "pricing_input_per_m": 2, "pricing_output_per_m": 10, "openness": "Proprietary", "key_strength": "High-Throughput Autonomous Software Engineering", "value_index": 68.9, "context_window": "1000K", "delta_prev": "-0.3", "trend": "down", "release_date": "2026-07-29", "benchmarks": { "gdpval": 56.5, "epoch_eci": 55.8, "gpqa_diamond": 86.4, "hle": 51.2, "frontier_math": 28.1, "arc_agi_3": 78.9, "swe_bench_verified": 85.2, "terminal_bench": 89.8, "live_code_bench": 84.6, "scicode": 79.2, "osworld": 72.4, "browse_comp": 81.5, "automation_bench": 85.8, "gaia": 80.4, "vectara_hallucination": 1.7, "facts_grounding": 96.8, "uptime_pct": 99.95, "pricing_input_per_m": 2, "pricing_output_per_m": 10, "tokens_per_sec": 96.5, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1402, "openrouter_rank": 4 }, "replication_count": 6, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://www.anthropic.com/news/claude-sonnet-5-5", "official_docs_url": "https://docs.anthropic.com/claude/reference/sonnet-5-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-sonnet-5-5-max", "sha256": "1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f" }, { "rank": 8, "id": "deepseek-v4-1-pro-max", "name": "DeepSeek V4.1 Pro Max", "provider": "DeepSeek AI", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 88.5, "aa_intelligence": 50.2, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.43, "pricing_output_per_m": 0.87, "openness": "Open Weights", "key_strength": "MLA MoE Architecture · #1 Open STEM", "value_index": 87.2, "context_window": "1000K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-07-10", "benchmarks": { "gdpval": 50.8, "epoch_eci": 50.4, "gpqa_diamond": 83.2, "hle": 48.2, "frontier_math": 25.8, "arc_agi_3": 75.8, "swe_bench_verified": 77.4, "terminal_bench": 81.6, "live_code_bench": 78, "scicode": 73.5, "osworld": 66.8, "browse_comp": 76.4, "automation_bench": 79.8, "gaia": 75.4, "vectara_hallucination": 1.9, "facts_grounding": 95.6, "uptime_pct": 99.88, "pricing_input_per_m": 0.43, "pricing_output_per_m": 0.87, "tokens_per_sec": 82.4, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1382, "openrouter_rank": 8 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://github.com/deepseek-ai/DeepSeek-V4", "official_docs_url": "https://api-docs.deepseek.com/", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-v4-1-pro-max", "sha256": "0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a" }, { "rank": 9, "id": "muse-spark-1-3-max", "name": "Muse Spark 1.3 Max", "provider": "Meta AI", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 88.1, "aa_intelligence": 48.1, "confidence_interval_95": 0.7, "pricing_input_per_m": 0.92, "pricing_output_per_m": 3.5, "openness": "Open Weights", "key_strength": "MCP Atlas Tool Routing (88.1%) & Vibe Code", "value_index": 90.1, "context_window": "262K", "delta_prev": "+1.5", "trend": "up", "release_date": "2026-08-30", "benchmarks": { "gdpval": 48.6, "epoch_eci": 48.2, "gpqa_diamond": 81.6, "hle": 46.5, "frontier_math": 24.2, "arc_agi_3": 74.5, "swe_bench_verified": 75.8, "terminal_bench": 80.2, "live_code_bench": 76.5, "scicode": 72, "osworld": 65.4, "browse_comp": 75.1, "automation_bench": 88.1, "gaia": 74.2, "vectara_hallucination": 2.2, "facts_grounding": 95.1, "uptime_pct": 99.91, "pricing_input_per_m": 0.92, "pricing_output_per_m": 3.5, "tokens_per_sec": 94, "hf_license_audit": 90, "epoch_reproducibility": 82, "lmsys_arena_elo": 1374, "openrouter_rank": 9 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://ai.meta.com/research/muse-spark-1-3/", "official_docs_url": "https://github.com/facebookresearch/muse-spark", "eval_audit_url": "https://artificialanalysis.ai/models/muse-spark-1-3-max", "sha256": "9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f" }, { "rank": 10, "id": "claude-fable-5-1-max", "name": "Claude Fable 5.1 Max", "provider": "Anthropic", "tier": "Tier 1: Frontier ASI/AGI", "aki_score": 87.8, "aa_intelligence": 53.4, "confidence_interval_95": 0.6, "pricing_input_per_m": 10, "pricing_output_per_m": 50, "openness": "Proprietary", "key_strength": "Extended Coherence Narrative & Test-Writing (67%)", "value_index": 62.4, "context_window": "1000K", "delta_prev": "-0.1", "trend": "flat", "release_date": "2026-06-20", "benchmarks": { "gdpval": 53.9, "epoch_eci": 53.5, "gpqa_diamond": 85, "hle": 49, "frontier_math": 25.4, "arc_agi_3": 76.8, "swe_bench_verified": 81.5, "terminal_bench": 86.4, "live_code_bench": 82, "scicode": 77, "osworld": 68.2, "browse_comp": 77.8, "automation_bench": 83.4, "gaia": 78, "vectara_hallucination": 1.8, "facts_grounding": 96.4, "uptime_pct": 99.94, "pricing_input_per_m": 10, "pricing_output_per_m": 50, "tokens_per_sec": 62, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1391, "openrouter_rank": 10 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://www.anthropic.com/news/claude-fable-5-1", "official_docs_url": "https://docs.anthropic.com/claude/reference/fable-5-1", "eval_audit_url": "https://artificialanalysis.ai/models/claude-fable-5-1-max", "sha256": "8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e" }, { "rank": 11, "id": "glm-5-3-max", "name": "GLM-5.3 Max", "provider": "Zhipu AI", "tier": "Tier 2: High Autonomy", "aki_score": 87.4, "aa_intelligence": 44.8, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.9, "pricing_output_per_m": 2.86, "openness": "Open Weights", "key_strength": "AutoBench Work Automation Leader (48.8%)", "value_index": 84.3, "context_window": "1000K", "delta_prev": "+0.5", "trend": "up", "release_date": "2026-07-15", "benchmarks": { "gdpval": 45.4, "epoch_eci": 45, "gpqa_diamond": 80.8, "hle": 44.2, "frontier_math": 23, "arc_agi_3": 73.2, "swe_bench_verified": 73.8, "terminal_bench": 78.4, "live_code_bench": 74.8, "scicode": 70.5, "osworld": 64.2, "browse_comp": 74, "automation_bench": 84.8, "gaia": 73.5, "vectara_hallucination": 2.3, "facts_grounding": 94.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.9, "pricing_output_per_m": 2.86, "tokens_per_sec": 86.4, "hf_license_audit": 90, "epoch_reproducibility": 80, "lmsys_arena_elo": 1362, "openrouter_rank": 11 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://zhipuai.cn/glm-5-3", "official_docs_url": "https://open.bigmodel.cn/dev/api", "eval_audit_url": "https://artificialanalysis.ai/models/glm-5-3-max", "sha256": "7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d" }, { "rank": 12, "id": "mimo-v2-6-pro", "name": "MiMo-V2.6-Pro", "provider": "Xiaomi", "tier": "Tier 2: High Autonomy", "aki_score": 86.5, "aa_intelligence": 46.3, "confidence_interval_95": 0.6, "pricing_input_per_m": 0.6, "pricing_output_per_m": 2.4, "openness": "Open Weights", "key_strength": "V2.6 MoE 126B · Peak Value Efficiency (98.6)", "value_index": 98.6, "context_window": "131K", "delta_prev": "+0.7", "trend": "up", "release_date": "2026-08-05", "benchmarks": { "gdpval": 46.9, "epoch_eci": 46.5, "gpqa_diamond": 79.5, "hle": 42.8, "frontier_math": 22.1, "arc_agi_3": 72, "swe_bench_verified": 72.4, "terminal_bench": 77, "live_code_bench": 73.2, "scicode": 69.2, "osworld": 62.8, "browse_comp": 72.5, "automation_bench": 80.4, "gaia": 71.8, "vectara_hallucination": 2.4, "facts_grounding": 94.5, "uptime_pct": 99.88, "pricing_input_per_m": 0.6, "pricing_output_per_m": 2.4, "tokens_per_sec": 105, "hf_license_audit": 95, "epoch_reproducibility": 88, "lmsys_arena_elo": 1358, "openrouter_rank": 12 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://github.com/Xiaomi/MiMo-V2.6", "official_docs_url": "https://ai.mi.com/mimo-docs", "eval_audit_url": "https://artificialanalysis.ai/models/mimo-v2-6-pro", "sha256": "6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c" }, { "rank": 13, "id": "deepseek-v4-1-flash", "name": "DeepSeek V4.1 Flash", "provider": "DeepSeek AI", "tier": "Tier 2: High Autonomy", "aki_score": 86.2, "aa_intelligence": 39.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.09, "pricing_output_per_m": 0.18, "openness": "Open Weights", "key_strength": "Global Usage Volume Leader (33.6T Tokens)", "value_index": 91.2, "context_window": "1000K", "delta_prev": "+1.2", "trend": "up", "release_date": "2026-07-18", "benchmarks": { "gdpval": 40.1, "epoch_eci": 39.8, "gpqa_diamond": 78.2, "hle": 41.5, "frontier_math": 21, "arc_agi_3": 70.5, "swe_bench_verified": 70.8, "terminal_bench": 75.4, "live_code_bench": 71.5, "scicode": 67.8, "osworld": 61.2, "browse_comp": 70.8, "automation_bench": 78.5, "gaia": 70.2, "vectara_hallucination": 2.5, "facts_grounding": 94, "uptime_pct": 99.95, "pricing_input_per_m": 0.09, "pricing_output_per_m": 0.18, "tokens_per_sec": 165, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1352, "openrouter_rank": 13 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://github.com/deepseek-ai/DeepSeek-V4-Flash", "official_docs_url": "https://api-docs.deepseek.com/", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-v4-1-flash", "sha256": "5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b" }, { "rank": 14, "id": "gemini-3-8-flash", "name": "Gemini 3.8 Flash", "provider": "Google DeepMind", "tier": "Tier 2: High Autonomy", "aki_score": 85.5, "aa_intelligence": 40.9, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.75, "pricing_output_per_m": 3.75, "openness": "Proprietary", "key_strength": "Terminal-Bench 2.1 (89.4%) & Ultra-Fast Audio", "value_index": 88.9, "context_window": "1048K", "delta_prev": "+1.0", "trend": "up", "release_date": "2026-06-25", "benchmarks": { "gdpval": 41.5, "epoch_eci": 41.1, "gpqa_diamond": 77, "hle": 40.2, "frontier_math": 20.2, "arc_agi_3": 69.4, "swe_bench_verified": 69.4, "terminal_bench": 89.4, "live_code_bench": 70.1, "scicode": 66.4, "osworld": 60.1, "browse_comp": 69.5, "automation_bench": 77.2, "gaia": 68.9, "vectara_hallucination": 2.6, "facts_grounding": 93.8, "uptime_pct": 99.98, "pricing_input_per_m": 0.75, "pricing_output_per_m": 3.75, "tokens_per_sec": 142, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1345, "openrouter_rank": 14 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://deepmind.google/technologies/gemini/flash/", "official_docs_url": "https://ai.google.dev/gemini-api/docs/models/gemini-3-8-flash", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-3-8-flash", "sha256": "4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d5e6f7a8b9c0d1e2f3a4b5c" }, { "rank": 15, "id": "grok-4-7-colossus", "name": "Grok 4.7 Colossus", "provider": "xAI", "tier": "Tier 2: High Autonomy", "aki_score": 85.1, "aa_intelligence": 46.4, "confidence_interval_95": 0.7, "pricing_input_per_m": 2, "pricing_output_per_m": 6, "openness": "Proprietary", "key_strength": "555k GPU Cluster Scale · Deep Live Web Grounding", "value_index": 54.2, "context_window": "2000K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-08-10", "benchmarks": { "gdpval": 47, "epoch_eci": 46.6, "gpqa_diamond": 78.4, "hle": 41.2, "frontier_math": 21.4, "arc_agi_3": 70.8, "swe_bench_verified": 71.2, "terminal_bench": 76.2, "live_code_bench": 72, "scicode": 67.5, "osworld": 62.4, "browse_comp": 84.5, "automation_bench": 78, "gaia": 71, "vectara_hallucination": 2.7, "facts_grounding": 94.2, "uptime_pct": 99.85, "pricing_input_per_m": 2, "pricing_output_per_m": 6, "tokens_per_sec": 88, "hf_license_audit": 10, "epoch_reproducibility": 25, "lmsys_arena_elo": 1360, "openrouter_rank": 15 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.02, "primary_source_url": "https://x.ai/grok-4", "official_docs_url": "https://docs.x.ai/", "eval_audit_url": "https://artificialanalysis.ai/models/grok-4-7-colossus", "sha256": "3c4d5e6f7a8b9c0d1e2f3a4b5c6d7e8f9a0b1c2d3e4f5a6b7c8d9e0f1a2b3c4d" }, { "rank": 16, "id": "claude-haiku-5-5", "name": "Claude Haiku 5.5", "provider": "Anthropic", "tier": "Tier 2: High Autonomy", "aki_score": 84.7, "aa_intelligence": 48.2, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.8, "pricing_output_per_m": 4, "openness": "Proprietary", "key_strength": "High-Speed Sub-Agent Worker", "value_index": 23.8, "context_window": "500K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-07-29", "benchmarks": { "gdpval": 48.7, "epoch_eci": 48.3, "gpqa_diamond": 79.1, "hle": 51.6, "frontier_math": 29.1, "arc_agi_3": 74.1, "swe_bench_verified": 77.6, "terminal_bench": 83.4, "live_code_bench": 76.7, "scicode": 73.4, "osworld": 65.8, "browse_comp": 73.4, "automation_bench": 76.7, "gaia": 72.5, "vectara_hallucination": 2, "facts_grounding": 95.6, "uptime_pct": 99.95, "pricing_input_per_m": 0.8, "pricing_output_per_m": 4, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1382, "openrouter_rank": 16 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-haiku-5-5", "official_docs_url": "https://aki1k.com/models/claude-haiku-5-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-haiku-5-5", "sha256": "sha256_2026_16_claude-haiku-5-5" }, { "rank": 17, "id": "gpt-5-6-sol", "name": "GPT-5.6 Sol", "provider": "OpenAI", "tier": "Tier 2: High Autonomy", "aki_score": 84.3, "aa_intelligence": 47.9, "confidence_interval_95": 0.4, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "openness": "Proprietary", "key_strength": "Speculative Frontier Coherence", "value_index": 15.6, "context_window": "500K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-07-15", "benchmarks": { "gdpval": 48.4, "epoch_eci": 48, "gpqa_diamond": 78.9, "hle": 51.4, "frontier_math": 29, "arc_agi_3": 73.9, "swe_bench_verified": 77.3, "terminal_bench": 83.2, "live_code_bench": 76.5, "scicode": 73.1, "osworld": 65.6, "browse_comp": 73.1, "automation_bench": 76.5, "gaia": 72.3, "vectara_hallucination": 2, "facts_grounding": 95.5, "uptime_pct": 99.95, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1381, "openrouter_rank": 17 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-5-6-sol", "official_docs_url": "https://aki1k.com/models/gpt-5-6-sol", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-5-6-sol", "sha256": "sha256_2026_17_gpt-5-6-sol" }, { "rank": 18, "id": "gemini-3-5-pro", "name": "Gemini 3.5 Pro", "provider": "Google DeepMind", "tier": "Tier 2: High Autonomy", "aki_score": 83.9, "aa_intelligence": 47.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 2, "pricing_output_per_m": 8, "openness": "Proprietary", "key_strength": "Multimodal Context Synthesis", "value_index": 11.6, "context_window": "2000K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-06-18", "benchmarks": { "gdpval": 48, "epoch_eci": 47.6, "gpqa_diamond": 78.7, "hle": 51.1, "frontier_math": 28.8, "arc_agi_3": 73.7, "swe_bench_verified": 77.1, "terminal_bench": 82.9, "live_code_bench": 76.2, "scicode": 72.9, "osworld": 65.3, "browse_comp": 72.9, "automation_bench": 76.2, "gaia": 72, "vectara_hallucination": 2.1, "facts_grounding": 95.5, "uptime_pct": 99.95, "pricing_input_per_m": 2, "pricing_output_per_m": 8, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1380, "openrouter_rank": 18 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-3-5-pro", "official_docs_url": "https://aki1k.com/models/gemini-3-5-pro", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-3-5-pro", "sha256": "sha256_2026_18_gemini-3-5-pro" }, { "rank": 19, "id": "sakana-fugu-ultra", "name": "Sakana Fugu Ultra", "provider": "Sakana AI", "tier": "Tier 2: High Autonomy", "aki_score": 83.2, "aa_intelligence": 46.8, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "openness": "Open Weights", "key_strength": "Evolutionary Model Merging", "value_index": 46.1, "context_window": "256K", "delta_prev": "+1.1", "trend": "up", "release_date": "2026-08-02", "benchmarks": { "gdpval": 47.3, "epoch_eci": 46.9, "gpqa_diamond": 78.3, "hle": 50.8, "frontier_math": 28.6, "arc_agi_3": 73.3, "swe_bench_verified": 76.6, "terminal_bench": 82.4, "live_code_bench": 75.7, "scicode": 72.4, "osworld": 64.9, "browse_comp": 72.4, "automation_bench": 75.7, "gaia": 71.6, "vectara_hallucination": 2.1, "facts_grounding": 95.4, "uptime_pct": 99.95, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1379, "openrouter_rank": 19 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/sakana-fugu-ultra", "official_docs_url": "https://aki1k.com/models/sakana-fugu-ultra", "eval_audit_url": "https://artificialanalysis.ai/models/sakana-fugu-ultra", "sha256": "sha256_2026_19_sakana-fugu-ultr" }, { "rank": 20, "id": "tencent-hy3", "name": "Tencent Hy3", "provider": "Tencent Hunyuan", "tier": "Tier 2: High Autonomy", "aki_score": 82.8, "aa_intelligence": 46.2, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.7, "pricing_output_per_m": 2.8, "openness": "Proprietary", "key_strength": "East Asia Multimodal Logic", "value_index": 32.8, "context_window": "512K", "delta_prev": "+0.5", "trend": "up", "release_date": "2026-07-22", "benchmarks": { "gdpval": 46.7, "epoch_eci": 46.3, "gpqa_diamond": 78, "hle": 50.5, "frontier_math": 28.5, "arc_agi_3": 73, "swe_bench_verified": 76.3, "terminal_bench": 82.1, "live_code_bench": 75.5, "scicode": 72.2, "osworld": 64.7, "browse_comp": 72.2, "automation_bench": 75.5, "gaia": 71.3, "vectara_hallucination": 2.1, "facts_grounding": 95.3, "uptime_pct": 99.95, "pricing_input_per_m": 0.7, "pricing_output_per_m": 2.8, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1378, "openrouter_rank": 20 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/tencent-hy3", "official_docs_url": "https://aki1k.com/models/tencent-hy3", "eval_audit_url": "https://artificialanalysis.ai/models/tencent-hy3", "sha256": "sha256_2026_20_tencent-hy3" }, { "rank": 21, "id": "qwen-3-7-max", "name": "Qwen 3.7 Max", "provider": "Alibaba Cloud", "tier": "Tier 2: High Autonomy", "aki_score": 82.1, "aa_intelligence": 45.8, "confidence_interval_95": 0.4, "pricing_input_per_m": 1.2, "pricing_output_per_m": 4.8, "openness": "Open Weights", "key_strength": "Autonomous Tool Agent Loops", "value_index": 18.9, "context_window": "500K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-06-30", "benchmarks": { "gdpval": 46.3, "epoch_eci": 45.9, "gpqa_diamond": 77.6, "hle": 50.2, "frontier_math": 28.3, "arc_agi_3": 72.6, "swe_bench_verified": 75.8, "terminal_bench": 81.6, "live_code_bench": 75, "scicode": 71.7, "osworld": 64.3, "browse_comp": 71.7, "automation_bench": 75, "gaia": 70.9, "vectara_hallucination": 2.1, "facts_grounding": 95.3, "uptime_pct": 99.95, "pricing_input_per_m": 1.2, "pricing_output_per_m": 4.8, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1377, "openrouter_rank": 21 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/qwen-3-7-max", "official_docs_url": "https://aki1k.com/models/qwen-3-7-max", "eval_audit_url": "https://artificialanalysis.ai/models/qwen-3-7-max", "sha256": "sha256_2026_21_qwen-3-7-max" }, { "rank": 22, "id": "grok-4-20", "name": "Grok 4.20", "provider": "xAI", "tier": "Tier 2: High Autonomy", "aki_score": 81.6, "aa_intelligence": 45.1, "confidence_interval_95": 0.6, "pricing_input_per_m": 1.5, "pricing_output_per_m": 5, "openness": "Proprietary", "key_strength": "Real-Time Sentiment & Grounding", "value_index": 17.8, "context_window": "1000K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-06-10", "benchmarks": { "gdpval": 45.6, "epoch_eci": 45.2, "gpqa_diamond": 77.3, "hle": 49.9, "frontier_math": 28.1, "arc_agi_3": 72.3, "swe_bench_verified": 75.5, "terminal_bench": 81.3, "live_code_bench": 74.7, "scicode": 71.4, "osworld": 64, "browse_comp": 71.4, "automation_bench": 74.7, "gaia": 70.6, "vectara_hallucination": 2.1, "facts_grounding": 95.2, "uptime_pct": 99.95, "pricing_input_per_m": 1.5, "pricing_output_per_m": 5, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1375, "openrouter_rank": 22 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/grok-4-20", "official_docs_url": "https://aki1k.com/models/grok-4-20", "eval_audit_url": "https://artificialanalysis.ai/models/grok-4-20", "sha256": "sha256_2026_22_grok-4-20" }, { "rank": 23, "id": "claude-opus-5", "name": "Claude Opus 5", "provider": "Anthropic", "tier": "Tier 2: High Autonomy", "aki_score": 81.2, "aa_intelligence": 44.7, "confidence_interval_95": 0.4, "pricing_input_per_m": 5, "pricing_output_per_m": 25, "openness": "Proprietary", "key_strength": "Complex Contract & Policy Synthesis", "value_index": 3.7, "context_window": "500K", "delta_prev": "-0.4", "trend": "down", "release_date": "2026-05-14", "benchmarks": { "gdpval": 45.2, "epoch_eci": 44.8, "gpqa_diamond": 77.1, "hle": 49.7, "frontier_math": 28, "arc_agi_3": 72.1, "swe_bench_verified": 75.2, "terminal_bench": 81, "live_code_bench": 74.4, "scicode": 71.2, "osworld": 63.7, "browse_comp": 71.2, "automation_bench": 74.4, "gaia": 70.3, "vectara_hallucination": 2.2, "facts_grounding": 95.2, "uptime_pct": 99.95, "pricing_input_per_m": 5, "pricing_output_per_m": 25, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1375, "openrouter_rank": 23 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-opus-5", "official_docs_url": "https://aki1k.com/models/claude-opus-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-opus-5", "sha256": "sha256_2026_23_claude-opus-5" }, { "rank": 24, "id": "claude-sonnet-5", "name": "Claude Sonnet 5", "provider": "Anthropic", "tier": "Tier 2: High Autonomy", "aki_score": 80.9, "aa_intelligence": 44.3, "confidence_interval_95": 0.4, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.5, "openness": "Proprietary", "key_strength": "Full-Stack Multi-File Refactoring", "value_index": 12, "context_window": "500K", "delta_prev": "-0.2", "trend": "down", "release_date": "2026-05-14", "benchmarks": { "gdpval": 44.8, "epoch_eci": 44.4, "gpqa_diamond": 76.9, "hle": 49.5, "frontier_math": 27.9, "arc_agi_3": 71.9, "swe_bench_verified": 75, "terminal_bench": 80.8, "live_code_bench": 74.2, "scicode": 71, "osworld": 63.5, "browse_comp": 71, "automation_bench": 74.2, "gaia": 70.2, "vectara_hallucination": 2.2, "facts_grounding": 95.1, "uptime_pct": 99.95, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.5, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1374, "openrouter_rank": 24 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-sonnet-5", "official_docs_url": "https://aki1k.com/models/claude-sonnet-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-sonnet-5", "sha256": "sha256_2026_24_claude-sonnet-5" }, { "rank": 25, "id": "gemini-3-5-flash", "name": "Gemini 3.5 Flash", "provider": "Google DeepMind", "tier": "Tier 2: High Autonomy", "aki_score": 80.4, "aa_intelligence": 43.8, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "openness": "Proprietary", "key_strength": "Low-Latency Agent Telemetry", "value_index": 44.5, "context_window": "1000K", "delta_prev": "+0.6", "trend": "up", "release_date": "2026-05-20", "benchmarks": { "gdpval": 44.3, "epoch_eci": 43.9, "gpqa_diamond": 76.6, "hle": 49.2, "frontier_math": 27.7, "arc_agi_3": 71.6, "swe_bench_verified": 74.7, "terminal_bench": 80.5, "live_code_bench": 73.9, "scicode": 70.7, "osworld": 63.2, "browse_comp": 70.7, "automation_bench": 73.9, "gaia": 69.8, "vectara_hallucination": 2.2, "facts_grounding": 95, "uptime_pct": 99.95, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1373, "openrouter_rank": 25 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-3-5-flash", "official_docs_url": "https://aki1k.com/models/gemini-3-5-flash", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-3-5-flash", "sha256": "sha256_2026_25_gemini-3-5-flash" }, { "rank": 26, "id": "deepseek-v4-pro", "name": "DeepSeek V4 Pro", "provider": "DeepSeek AI", "tier": "Tier 2: High Autonomy", "aki_score": 79.8, "aa_intelligence": 43.2, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.35, "pricing_output_per_m": 0.7, "openness": "Open Weights", "key_strength": "Dual-Pipe Attention Efficiency", "value_index": 117.3, "context_window": "500K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-05-01", "benchmarks": { "gdpval": 43.7, "epoch_eci": 43.3, "gpqa_diamond": 76.3, "hle": 48.9, "frontier_math": 27.5, "arc_agi_3": 71.3, "swe_bench_verified": 74.3, "terminal_bench": 80.1, "live_code_bench": 73.5, "scicode": 70.3, "osworld": 62.9, "browse_comp": 70.3, "automation_bench": 73.5, "gaia": 69.5, "vectara_hallucination": 2.2, "facts_grounding": 95, "uptime_pct": 99.95, "pricing_input_per_m": 0.35, "pricing_output_per_m": 0.7, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1372, "openrouter_rank": 26 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-v4-pro", "official_docs_url": "https://aki1k.com/models/deepseek-v4-pro", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-v4-pro", "sha256": "sha256_2026_26_deepseek-v4-pro" }, { "rank": 27, "id": "deepseek-v4-flash", "name": "DeepSeek V4 Flash", "provider": "DeepSeek AI", "tier": "Tier 2: High Autonomy", "aki_score": 79.1, "aa_intelligence": 42.6, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.08, "pricing_output_per_m": 0.16, "openness": "Open Weights", "key_strength": "High-Throughput Batch Inference", "value_index": 508.5, "context_window": "500K", "delta_prev": "+0.8", "trend": "up", "release_date": "2026-05-08", "benchmarks": { "gdpval": 43.1, "epoch_eci": 42.7, "gpqa_diamond": 75.9, "hle": 48.5, "frontier_math": 27.3, "arc_agi_3": 70.9, "swe_bench_verified": 73.8, "terminal_bench": 79.6, "live_code_bench": 73, "scicode": 69.8, "osworld": 62.5, "browse_comp": 69.8, "automation_bench": 73, "gaia": 69, "vectara_hallucination": 2.2, "facts_grounding": 94.9, "uptime_pct": 99.95, "pricing_input_per_m": 0.08, "pricing_output_per_m": 0.16, "tokens_per_sec": 80, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1370, "openrouter_rank": 27 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-v4-flash", "official_docs_url": "https://aki1k.com/models/deepseek-v4-flash", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-v4-flash", "sha256": "sha256_2026_27_deepseek-v4-flas" }, { "rank": 28, "id": "gpt-5-5", "name": "GPT-5.5", "provider": "OpenAI", "tier": "Tier 2: High Autonomy", "aki_score": 78.6, "aa_intelligence": 42.1, "confidence_interval_95": 0.4, "pricing_input_per_m": 3, "pricing_output_per_m": 12, "openness": "Proprietary", "key_strength": "General Multimodal Operations", "value_index": 7.3, "context_window": "500K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-04-12", "benchmarks": { "gdpval": 42.6, "epoch_eci": 42.2, "gpqa_diamond": 75.6, "hle": 48.2, "frontier_math": 27.2, "arc_agi_3": 70.6, "swe_bench_verified": 73.4, "terminal_bench": 79.2, "live_code_bench": 72.7, "scicode": 69.5, "osworld": 62.2, "browse_comp": 69.5, "automation_bench": 72.7, "gaia": 68.7, "vectara_hallucination": 2.2, "facts_grounding": 94.8, "uptime_pct": 99.95, "pricing_input_per_m": 3, "pricing_output_per_m": 12, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1369, "openrouter_rank": 28 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-5-5", "official_docs_url": "https://aki1k.com/models/gpt-5-5", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-5-5", "sha256": "sha256_2026_28_gpt-5-5" }, { "rank": 29, "id": "claude-opus-4-8", "name": "Claude Opus 4.8", "provider": "Anthropic", "tier": "Tier 2: High Autonomy", "aki_score": 77.9, "aa_intelligence": 41.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 6, "pricing_output_per_m": 30, "openness": "Proprietary", "key_strength": "Long-Context Actuarial Reasoning", "value_index": 2.9, "context_window": "500K", "delta_prev": "-0.3", "trend": "down", "release_date": "2026-03-25", "benchmarks": { "gdpval": 42, "epoch_eci": 41.6, "gpqa_diamond": 75.2, "hle": 47.8, "frontier_math": 26.9, "arc_agi_3": 70.2, "swe_bench_verified": 73, "terminal_bench": 78.8, "live_code_bench": 72.2, "scicode": 69.1, "osworld": 61.7, "browse_comp": 69.1, "automation_bench": 72.2, "gaia": 68.3, "vectara_hallucination": 2.3, "facts_grounding": 94.7, "uptime_pct": 99.95, "pricing_input_per_m": 6, "pricing_output_per_m": 30, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1367, "openrouter_rank": 29 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-opus-4-8", "official_docs_url": "https://aki1k.com/models/claude-opus-4-8", "eval_audit_url": "https://artificialanalysis.ai/models/claude-opus-4-8", "sha256": "sha256_2026_29_claude-opus-4-8" }, { "rank": 30, "id": "glm-5-2", "name": "GLM-5.2", "provider": "Zhipu AI", "tier": "Tier 2: High Autonomy", "aki_score": 76.5, "aa_intelligence": 40.8, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "openness": "Open Weights", "key_strength": "Enterprise Code & Knowledge Work", "value_index": 34.4, "context_window": "500K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-04-18", "benchmarks": { "gdpval": 41.3, "epoch_eci": 40.9, "gpqa_diamond": 74.4, "hle": 47.1, "frontier_math": 26.5, "arc_agi_3": 69.4, "swe_bench_verified": 72, "terminal_bench": 77.8, "live_code_bench": 71.3, "scicode": 68.2, "osworld": 60.9, "browse_comp": 68.2, "automation_bench": 71.3, "gaia": 67.4, "vectara_hallucination": 2.3, "facts_grounding": 94.6, "uptime_pct": 99.95, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1364, "openrouter_rank": 30 }, "replication_count": 5, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/glm-5-2", "official_docs_url": "https://aki1k.com/models/glm-5-2", "eval_audit_url": "https://artificialanalysis.ai/models/glm-5-2", "sha256": "sha256_2026_30_glm-5-2" }, { "rank": 31, "id": "gpt-5-4", "name": "GPT-5.4", "provider": "OpenAI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 74.8, "aa_intelligence": 39.8, "confidence_interval_95": 0.4, "pricing_input_per_m": 2.5, "pricing_output_per_m": 10, "openness": "Proprietary", "key_strength": "Reliable Enterprise Pipeline Routing", "value_index": 8.3, "context_window": "256K", "delta_prev": "-0.2", "trend": "down", "release_date": "2026-02-15", "benchmarks": { "gdpval": 40.3, "epoch_eci": 39.9, "gpqa_diamond": 73.4, "hle": 46.1, "frontier_math": 25.9, "arc_agi_3": 68.4, "swe_bench_verified": 70.9, "terminal_bench": 76.6, "live_code_bench": 70.1, "scicode": 67.1, "osworld": 59.9, "browse_comp": 67.1, "automation_bench": 70.1, "gaia": 66.4, "vectara_hallucination": 2.4, "facts_grounding": 94.3, "uptime_pct": 99.95, "pricing_input_per_m": 2.5, "pricing_output_per_m": 10, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1361, "openrouter_rank": 31 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-5-4", "official_docs_url": "https://aki1k.com/models/gpt-5-4", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-5-4", "sha256": "sha256_2026_31_gpt-5-4" }, { "rank": 32, "id": "claude-sonnet-4-6", "name": "Claude Sonnet 4.6", "provider": "Anthropic", "tier": "Tier 3: Enterprise Reliable", "aki_score": 74.2, "aa_intelligence": 39.2, "confidence_interval_95": 0.4, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "openness": "Proprietary", "key_strength": "Codebase Migration & Testing", "value_index": 13.7, "context_window": "256K", "delta_prev": "-0.3", "trend": "down", "release_date": "2026-02-20", "benchmarks": { "gdpval": 39.7, "epoch_eci": 39.3, "gpqa_diamond": 73, "hle": 45.8, "frontier_math": 25.7, "arc_agi_3": 68, "swe_bench_verified": 70.5, "terminal_bench": 76.2, "live_code_bench": 69.7, "scicode": 66.7, "osworld": 59.5, "browse_comp": 66.7, "automation_bench": 69.7, "gaia": 66, "vectara_hallucination": 2.4, "facts_grounding": 94.3, "uptime_pct": 99.95, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1360, "openrouter_rank": 32 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-sonnet-4-6", "official_docs_url": "https://aki1k.com/models/claude-sonnet-4-6", "eval_audit_url": "https://artificialanalysis.ai/models/claude-sonnet-4-6", "sha256": "sha256_2026_32_claude-sonnet-4-" }, { "rank": 33, "id": "gemini-3-1-pro", "name": "Gemini 3.1 Pro", "provider": "Google DeepMind", "tier": "Tier 3: Enterprise Reliable", "aki_score": 73.9, "aa_intelligence": 38.9, "confidence_interval_95": 0.3, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.2, "openness": "Proprietary", "key_strength": "Document OCR & Data Extraction", "value_index": 11.4, "context_window": "1000K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-03-01", "benchmarks": { "gdpval": 39.4, "epoch_eci": 39, "gpqa_diamond": 72.9, "hle": 45.6, "frontier_math": 25.6, "arc_agi_3": 67.9, "swe_bench_verified": 70.3, "terminal_bench": 76, "live_code_bench": 69.5, "scicode": 66.6, "osworld": 59.3, "browse_comp": 66.6, "automation_bench": 69.5, "gaia": 65.8, "vectara_hallucination": 2.4, "facts_grounding": 94.2, "uptime_pct": 99.95, "pricing_input_per_m": 1.8, "pricing_output_per_m": 7.2, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1359, "openrouter_rank": 33 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-3-1-pro", "official_docs_url": "https://aki1k.com/models/gemini-3-1-pro", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-3-1-pro", "sha256": "sha256_2026_33_gemini-3-1-pro" }, { "rank": 34, "id": "qwen-3-coder-32b", "name": "Qwen 3 Coder 32B", "provider": "Alibaba Cloud", "tier": "Tier 3: Enterprise Reliable", "aki_score": 73.4, "aa_intelligence": 38.4, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Open Weights", "key_strength": "Local IDE Assistant & Code Fill", "value_index": 176.2, "context_window": "128K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-03-12", "benchmarks": { "gdpval": 38.9, "epoch_eci": 38.5, "gpqa_diamond": 72.6, "hle": 45.4, "frontier_math": 25.5, "arc_agi_3": 67.6, "swe_bench_verified": 69.9, "terminal_bench": 75.6, "live_code_bench": 69.2, "scicode": 66.2, "osworld": 59, "browse_comp": 66.2, "automation_bench": 69.2, "gaia": 65.5, "vectara_hallucination": 2.4, "facts_grounding": 94.2, "uptime_pct": 99.95, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1358, "openrouter_rank": 34 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/qwen-3-coder-32b", "official_docs_url": "https://aki1k.com/models/qwen-3-coder-32b", "eval_audit_url": "https://artificialanalysis.ai/models/qwen-3-coder-32b", "sha256": "sha256_2026_34_qwen-3-coder-32b" }, { "rank": 35, "id": "deepseek-coder-v3", "name": "DeepSeek Coder V3", "provider": "DeepSeek AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 72.8, "aa_intelligence": 37.9, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.12, "pricing_output_per_m": 0.24, "openness": "Open Weights", "key_strength": "Deterministic Syntax & Refactoring", "value_index": 312, "context_window": "128K", "delta_prev": "+0.2", "trend": "flat", "release_date": "2026-03-15", "benchmarks": { "gdpval": 38.4, "epoch_eci": 38, "gpqa_diamond": 72.2, "hle": 45, "frontier_math": 25.3, "arc_agi_3": 67.2, "swe_bench_verified": 69.5, "terminal_bench": 75.2, "live_code_bench": 68.8, "scicode": 65.9, "osworld": 58.7, "browse_comp": 65.9, "automation_bench": 68.8, "gaia": 65.1, "vectara_hallucination": 2.5, "facts_grounding": 94.1, "uptime_pct": 99.95, "pricing_input_per_m": 0.12, "pricing_output_per_m": 0.24, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1357, "openrouter_rank": 35 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-coder-v3", "official_docs_url": "https://aki1k.com/models/deepseek-coder-v3", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-coder-v3", "sha256": "sha256_2026_35_deepseek-coder-v" }, { "rank": 36, "id": "llama-4-maverick", "name": "Llama 4 Maverick", "provider": "Meta AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 72.1, "aa_intelligence": 37.2, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.2, "openness": "Open Weights", "key_strength": "High-Throughput Self-Hosted Core", "value_index": 64.9, "context_window": "256K", "delta_prev": "+0.6", "trend": "up", "release_date": "2026-04-05", "benchmarks": { "gdpval": 37.7, "epoch_eci": 37.3, "gpqa_diamond": 71.8, "hle": 44.7, "frontier_math": 25.1, "arc_agi_3": 66.8, "swe_bench_verified": 69, "terminal_bench": 74.7, "live_code_bench": 68.3, "scicode": 65.4, "osworld": 58.3, "browse_comp": 65.4, "automation_bench": 68.3, "gaia": 64.7, "vectara_hallucination": 2.5, "facts_grounding": 94, "uptime_pct": 99.95, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.2, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1355, "openrouter_rank": 36 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-4-maverick", "official_docs_url": "https://aki1k.com/models/llama-4-maverick", "eval_audit_url": "https://artificialanalysis.ai/models/llama-4-maverick", "sha256": "sha256_2026_36_llama-4-maverick" }, { "rank": 37, "id": "mistral-large-3-1", "name": "Mistral Large 3.1", "provider": "Mistral AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 71.6, "aa_intelligence": 36.8, "confidence_interval_95": 0.4, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "openness": "Open Weights", "key_strength": "Multilingual EU Compliance & Reasoning", "value_index": 25.8, "context_window": "256K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-04-22", "benchmarks": { "gdpval": 37.3, "epoch_eci": 36.9, "gpqa_diamond": 71.5, "hle": 44.4, "frontier_math": 24.9, "arc_agi_3": 66.5, "swe_bench_verified": 68.7, "terminal_bench": 74.4, "live_code_bench": 68, "scicode": 65.1, "osworld": 58, "browse_comp": 65.1, "automation_bench": 68, "gaia": 64.4, "vectara_hallucination": 2.5, "facts_grounding": 94, "uptime_pct": 99.95, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "tokens_per_sec": 80, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1354, "openrouter_rank": 37 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/mistral-large-3-1", "official_docs_url": "https://aki1k.com/models/mistral-large-3-1", "eval_audit_url": "https://artificialanalysis.ai/models/mistral-large-3-1", "sha256": "sha256_2026_37_mistral-large-3-" }, { "rank": 38, "id": "gemma-3-27b", "name": "Gemma 3 27B", "provider": "Google DeepMind", "tier": "Tier 3: Enterprise Reliable", "aki_score": 71, "aa_intelligence": 36.2, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "openness": "Open Weights", "key_strength": "Lightweight Local Workstation AI", "value_index": 255.6, "context_window": "128K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-04-10", "benchmarks": { "gdpval": 36.7, "epoch_eci": 36.3, "gpqa_diamond": 71.2, "hle": 44, "frontier_math": 24.7, "arc_agi_3": 66.2, "swe_bench_verified": 68.3, "terminal_bench": 74, "live_code_bench": 67.6, "scicode": 64.7, "osworld": 57.6, "browse_comp": 64.7, "automation_bench": 67.6, "gaia": 64, "vectara_hallucination": 2.5, "facts_grounding": 93.9, "uptime_pct": 99.95, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "tokens_per_sec": 100, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1353, "openrouter_rank": 38 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemma-3-27b", "official_docs_url": "https://aki1k.com/models/gemma-3-27b", "eval_audit_url": "https://artificialanalysis.ai/models/gemma-3-27b", "sha256": "sha256_2026_38_gemma-3-27b" }, { "rank": 39, "id": "claude-haiku-4-5", "name": "Claude Haiku 4.5", "provider": "Anthropic", "tier": "Tier 3: Enterprise Reliable", "aki_score": 70.5, "aa_intelligence": 35.7, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "openness": "Proprietary", "key_strength": "Fast Classification & Moderation", "value_index": 39, "context_window": "256K", "delta_prev": "-0.4", "trend": "down", "release_date": "2026-01-18", "benchmarks": { "gdpval": 36.2, "epoch_eci": 35.8, "gpqa_diamond": 70.9, "hle": 43.8, "frontier_math": 24.6, "arc_agi_3": 65.9, "swe_bench_verified": 67.9, "terminal_bench": 73.6, "live_code_bench": 67.2, "scicode": 64.4, "osworld": 57.3, "browse_comp": 64.4, "automation_bench": 67.2, "gaia": 63.7, "vectara_hallucination": 2.5, "facts_grounding": 93.8, "uptime_pct": 99.95, "pricing_input_per_m": 0.5, "pricing_output_per_m": 2, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1352, "openrouter_rank": 39 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-haiku-4-5", "official_docs_url": "https://aki1k.com/models/claude-haiku-4-5", "eval_audit_url": "https://artificialanalysis.ai/models/claude-haiku-4-5", "sha256": "sha256_2026_39_claude-haiku-4-5" }, { "rank": 40, "id": "gpt-5-3-codex", "name": "GPT-5.3 Codex", "provider": "OpenAI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 69.9, "aa_intelligence": 35.1, "confidence_interval_95": 0.4, "pricing_input_per_m": 1.2, "pricing_output_per_m": 4.8, "openness": "Proprietary", "key_strength": "Sandboxed Script Generation", "value_index": 16.1, "context_window": "128K", "delta_prev": "-0.2", "trend": "down", "release_date": "2026-01-25", "benchmarks": { "gdpval": 35.6, "epoch_eci": 35.2, "gpqa_diamond": 70.5, "hle": 43.4, "frontier_math": 24.4, "arc_agi_3": 65.5, "swe_bench_verified": 67.5, "terminal_bench": 73.2, "live_code_bench": 66.8, "scicode": 64, "osworld": 56.9, "browse_comp": 64, "automation_bench": 66.8, "gaia": 63.3, "vectara_hallucination": 2.6, "facts_grounding": 93.7, "uptime_pct": 99.95, "pricing_input_per_m": 1.2, "pricing_output_per_m": 4.8, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1350, "openrouter_rank": 40 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-5-3-codex", "official_docs_url": "https://aki1k.com/models/gpt-5-3-codex", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-5-3-codex", "sha256": "sha256_2026_40_gpt-5-3-codex" }, { "rank": 41, "id": "nova-pro-2", "name": "Nova Pro 2", "provider": "Amazon AWS", "tier": "Tier 3: Enterprise Reliable", "aki_score": 69.2, "aa_intelligence": 34.6, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.8, "pricing_output_per_m": 3.2, "openness": "Proprietary", "key_strength": "AWS Bedrock Enterprise Workloads", "value_index": 24, "context_window": "300K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-03-20", "benchmarks": { "gdpval": 35.1, "epoch_eci": 34.7, "gpqa_diamond": 70.1, "hle": 43.1, "frontier_math": 24.1, "arc_agi_3": 65.1, "swe_bench_verified": 67.1, "terminal_bench": 72.7, "live_code_bench": 66.4, "scicode": 63.6, "osworld": 56.5, "browse_comp": 63.6, "automation_bench": 66.4, "gaia": 62.9, "vectara_hallucination": 2.6, "facts_grounding": 93.7, "uptime_pct": 99.95, "pricing_input_per_m": 0.8, "pricing_output_per_m": 3.2, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1349, "openrouter_rank": 41 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/nova-pro-2", "official_docs_url": "https://aki1k.com/models/nova-pro-2", "eval_audit_url": "https://artificialanalysis.ai/models/nova-pro-2", "sha256": "sha256_2026_41_nova-pro-2" }, { "rank": 42, "id": "command-r-2-0", "name": "Command R+ 2.0", "provider": "Cohere", "tier": "Tier 3: Enterprise Reliable", "aki_score": 68.7, "aa_intelligence": 34, "confidence_interval_95": 0.5, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "openness": "Proprietary", "key_strength": "Multilingual Enterprise RAG & Citations", "value_index": 12.7, "context_window": "256K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-02-28", "benchmarks": { "gdpval": 34.5, "epoch_eci": 34.1, "gpqa_diamond": 69.8, "hle": 42.8, "frontier_math": 24, "arc_agi_3": 64.8, "swe_bench_verified": 66.7, "terminal_bench": 72.4, "live_code_bench": 66, "scicode": 63.3, "osworld": 56.2, "browse_comp": 63.3, "automation_bench": 66, "gaia": 62.6, "vectara_hallucination": 2.6, "facts_grounding": 93.6, "uptime_pct": 99.95, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1348, "openrouter_rank": 42 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/command-r-2-0", "official_docs_url": "https://aki1k.com/models/command-r-2-0", "eval_audit_url": "https://artificialanalysis.ai/models/command-r-2-0", "sha256": "sha256_2026_42_command-r-2-0" }, { "rank": 43, "id": "yi-large-2-5", "name": "Yi-Large 2.5", "provider": "01.AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 67.8, "aa_intelligence": 33.4, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.9, "pricing_output_per_m": 2.7, "openness": "Proprietary", "key_strength": "Bicultural Reasoning & Summarization", "value_index": 27.1, "context_window": "256K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-01-30", "benchmarks": { "gdpval": 33.9, "epoch_eci": 33.5, "gpqa_diamond": 69.3, "hle": 42.3, "frontier_math": 23.7, "arc_agi_3": 64.3, "swe_bench_verified": 66.1, "terminal_bench": 71.8, "live_code_bench": 65.4, "scicode": 62.7, "osworld": 55.7, "browse_comp": 62.7, "automation_bench": 65.4, "gaia": 62, "vectara_hallucination": 2.6, "facts_grounding": 93.5, "uptime_pct": 99.95, "pricing_input_per_m": 0.9, "pricing_output_per_m": 2.7, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1346, "openrouter_rank": 43 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/yi-large-2-5", "official_docs_url": "https://aki1k.com/models/yi-large-2-5", "eval_audit_url": "https://artificialanalysis.ai/models/yi-large-2-5", "sha256": "sha256_2026_43_yi-large-2-5" }, { "rank": 44, "id": "jamba-1-8-large", "name": "Jamba 1.8 Large", "provider": "AI21 Labs", "tier": "Tier 3: Enterprise Reliable", "aki_score": 66.9, "aa_intelligence": 32.8, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.6, "pricing_output_per_m": 2.4, "openness": "Open Weights", "key_strength": "Hybrid SSM-Transformer Architecture", "value_index": 30.9, "context_window": "256K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-02-12", "benchmarks": { "gdpval": 33.3, "epoch_eci": 32.9, "gpqa_diamond": 68.8, "hle": 41.8, "frontier_math": 23.4, "arc_agi_3": 63.8, "swe_bench_verified": 65.5, "terminal_bench": 71.2, "live_code_bench": 64.8, "scicode": 62.1, "osworld": 55.1, "browse_comp": 62.1, "automation_bench": 64.8, "gaia": 61.5, "vectara_hallucination": 2.7, "facts_grounding": 93.4, "uptime_pct": 99.95, "pricing_input_per_m": 0.6, "pricing_output_per_m": 2.4, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1344, "openrouter_rank": 44 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/jamba-1-8-large", "official_docs_url": "https://aki1k.com/models/jamba-1-8-large", "eval_audit_url": "https://artificialanalysis.ai/models/jamba-1-8-large", "sha256": "sha256_2026_44_jamba-1-8-large" }, { "rank": 45, "id": "granite-4-0-20b", "name": "Granite 4.0 20B", "provider": "IBM Research", "tier": "Tier 3: Enterprise Reliable", "aki_score": 65.4, "aa_intelligence": 31.9, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Apache 2.0", "key_strength": "Enterprise Governance & Auditability", "value_index": 157, "context_window": "128K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-03-08", "benchmarks": { "gdpval": 32.4, "epoch_eci": 32, "gpqa_diamond": 67.9, "hle": 41, "frontier_math": 22.9, "arc_agi_3": 62.9, "swe_bench_verified": 64.5, "terminal_bench": 70.1, "live_code_bench": 63.8, "scicode": 61.2, "osworld": 54.2, "browse_comp": 61.2, "automation_bench": 63.8, "gaia": 60.5, "vectara_hallucination": 2.7, "facts_grounding": 93.2, "uptime_pct": 99.95, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 40, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1341, "openrouter_rank": 45 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/granite-4-0-20b", "official_docs_url": "https://aki1k.com/models/granite-4-0-20b", "eval_audit_url": "https://artificialanalysis.ai/models/granite-4-0-20b", "sha256": "sha256_2026_45_granite-4-0-20b" }, { "rank": 46, "id": "mistral-small-3", "name": "Mistral Small 3", "provider": "Mistral AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 65, "aa_intelligence": 31.4, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Open Weights", "key_strength": "Fast Local Orchestration", "value_index": 156, "context_window": "128K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-02-05", "benchmarks": { "gdpval": 31.9, "epoch_eci": 31.5, "gpqa_diamond": 67.7, "hle": 40.8, "frontier_math": 22.8, "arc_agi_3": 62.7, "swe_bench_verified": 64.2, "terminal_bench": 69.8, "live_code_bench": 63.6, "scicode": 61, "osworld": 54, "browse_comp": 61, "automation_bench": 63.6, "gaia": 60.3, "vectara_hallucination": 2.7, "facts_grounding": 93.1, "uptime_pct": 99.95, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1340, "openrouter_rank": 46 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/mistral-small-3", "official_docs_url": "https://aki1k.com/models/mistral-small-3", "eval_audit_url": "https://artificialanalysis.ai/models/mistral-small-3", "sha256": "sha256_2026_46_mistral-small-3" }, { "rank": 47, "id": "gpt-5", "name": "GPT-5", "provider": "OpenAI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 64.6, "aa_intelligence": 30.9, "confidence_interval_95": 0.4, "pricing_input_per_m": 2, "pricing_output_per_m": 8, "openness": "Proprietary", "key_strength": "Baseline Foundation Reasoning", "value_index": 8.9, "context_window": "128K", "delta_prev": "-0.5", "trend": "down", "release_date": "2025-11-20", "benchmarks": { "gdpval": 31.4, "epoch_eci": 31, "gpqa_diamond": 67.5, "hle": 40.5, "frontier_math": 22.7, "arc_agi_3": 62.5, "swe_bench_verified": 63.9, "terminal_bench": 69.6, "live_code_bench": 63.3, "scicode": 60.7, "osworld": 53.8, "browse_comp": 60.7, "automation_bench": 63.3, "gaia": 60.1, "vectara_hallucination": 2.7, "facts_grounding": 93.1, "uptime_pct": 99.95, "pricing_input_per_m": 2, "pricing_output_per_m": 8, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1339, "openrouter_rank": 47 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-5", "official_docs_url": "https://aki1k.com/models/gpt-5", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-5", "sha256": "sha256_2026_47_gpt-5" }, { "rank": 48, "id": "gemini-2-5-pro", "name": "Gemini 2.5 Pro", "provider": "Google DeepMind", "tier": "Tier 3: Enterprise Reliable", "aki_score": 64.2, "aa_intelligence": 30.4, "confidence_interval_95": 0.3, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "openness": "Proprietary", "key_strength": "Long-Context Baseline", "value_index": 11.9, "context_window": "1000K", "delta_prev": "-0.4", "trend": "down", "release_date": "2025-10-15", "benchmarks": { "gdpval": 30.9, "epoch_eci": 30.5, "gpqa_diamond": 67.2, "hle": 40.3, "frontier_math": 22.5, "arc_agi_3": 62.2, "swe_bench_verified": 63.7, "terminal_bench": 69.3, "live_code_bench": 63, "scicode": 60.4, "osworld": 53.5, "browse_comp": 60.4, "automation_bench": 63, "gaia": 59.8, "vectara_hallucination": 2.8, "facts_grounding": 93, "uptime_pct": 99.95, "pricing_input_per_m": 1.5, "pricing_output_per_m": 6, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1338, "openrouter_rank": 48 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-2-5-pro", "official_docs_url": "https://aki1k.com/models/gemini-2-5-pro", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-2-5-pro", "sha256": "sha256_2026_48_gemini-2-5-pro" }, { "rank": 49, "id": "gemini-2-5-flash", "name": "Gemini 2.5 Flash", "provider": "Google DeepMind", "tier": "Tier 3: Enterprise Reliable", "aki_score": 63.8, "aa_intelligence": 29.9, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.8, "openness": "Proprietary", "key_strength": "Multimodal Stream Processing", "value_index": 88.3, "context_window": "1000K", "delta_prev": "-0.3", "trend": "down", "release_date": "2025-10-22", "benchmarks": { "gdpval": 30.4, "epoch_eci": 30, "gpqa_diamond": 67, "hle": 40.1, "frontier_math": 22.4, "arc_agi_3": 62, "swe_bench_verified": 63.4, "terminal_bench": 69, "live_code_bench": 62.7, "scicode": 60.2, "osworld": 53.3, "browse_comp": 60.2, "automation_bench": 62.7, "gaia": 59.6, "vectara_hallucination": 2.8, "facts_grounding": 93, "uptime_pct": 99.95, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.8, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1337, "openrouter_rank": 49 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-2-5-flash", "official_docs_url": "https://aki1k.com/models/gemini-2-5-flash", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-2-5-flash", "sha256": "sha256_2026_49_gemini-2-5-flash" }, { "rank": 50, "id": "qwen-3-max", "name": "Qwen 3 Max", "provider": "Alibaba Cloud", "tier": "Tier 3: Enterprise Reliable", "aki_score": 63.4, "aa_intelligence": 29.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 1, "pricing_output_per_m": 3.5, "openness": "Open Weights", "key_strength": "General Agent Planning", "value_index": 19.8, "context_window": "256K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2026-01-15", "benchmarks": { "gdpval": 30, "epoch_eci": 29.6, "gpqa_diamond": 66.8, "hle": 39.9, "frontier_math": 22.3, "arc_agi_3": 61.8, "swe_bench_verified": 63.1, "terminal_bench": 68.7, "live_code_bench": 62.5, "scicode": 59.9, "osworld": 53, "browse_comp": 59.9, "automation_bench": 62.5, "gaia": 59.3, "vectara_hallucination": 2.8, "facts_grounding": 92.9, "uptime_pct": 99.95, "pricing_input_per_m": 1, "pricing_output_per_m": 3.5, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1336, "openrouter_rank": 50 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/qwen-3-max", "official_docs_url": "https://aki1k.com/models/qwen-3-max", "eval_audit_url": "https://artificialanalysis.ai/models/qwen-3-max", "sha256": "sha256_2026_50_qwen-3-max" }, { "rank": 51, "id": "o4-mini", "name": "o4-mini", "provider": "OpenAI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 63, "aa_intelligence": 29, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.6, "openness": "Proprietary", "key_strength": "Compact Mathematical Reasoning", "value_index": 43.6, "context_window": "128K", "delta_prev": "+0.2", "trend": "up", "release_date": "2026-02-10", "benchmarks": { "gdpval": 29.5, "epoch_eci": 29.1, "gpqa_diamond": 66.5, "hle": 39.6, "frontier_math": 22.2, "arc_agi_3": 61.5, "swe_bench_verified": 62.8, "terminal_bench": 68.5, "live_code_bench": 62.2, "scicode": 59.7, "osworld": 52.8, "browse_comp": 59.7, "automation_bench": 62.2, "gaia": 59.1, "vectara_hallucination": 2.8, "facts_grounding": 92.9, "uptime_pct": 99.85, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.6, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1335, "openrouter_rank": 51 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/o4-mini", "official_docs_url": "https://aki1k.com/models/o4-mini", "eval_audit_url": "https://artificialanalysis.ai/models/o4-mini", "sha256": "sha256_2026_51_o4-mini" }, { "rank": 52, "id": "llama-4-scout", "name": "Llama 4 Scout", "provider": "Meta AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 62.5, "aa_intelligence": 28.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Open Weights", "key_strength": "Efficient Open MoE Baseline", "value_index": 150, "context_window": "128K", "delta_prev": "+0.4", "trend": "up", "release_date": "2026-03-01", "benchmarks": { "gdpval": 29, "epoch_eci": 28.6, "gpqa_diamond": 66.2, "hle": 39.4, "frontier_math": 22, "arc_agi_3": 61.2, "swe_bench_verified": 62.5, "terminal_bench": 68.1, "live_code_bench": 61.9, "scicode": 59.4, "osworld": 52.5, "browse_comp": 59.4, "automation_bench": 61.9, "gaia": 58.8, "vectara_hallucination": 2.8, "facts_grounding": 92.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 80, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1334, "openrouter_rank": 52 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-4-scout", "official_docs_url": "https://aki1k.com/models/llama-4-scout", "eval_audit_url": "https://artificialanalysis.ai/models/llama-4-scout", "sha256": "sha256_2026_52_llama-4-scout" }, { "rank": 53, "id": "claude-mythos-preview-hold", "name": "Claude Mythos Preview [Hold]", "provider": "Anthropic", "tier": "Tier 3: Enterprise Reliable", "aki_score": 62, "aa_intelligence": 28, "confidence_interval_95": 0.7, "pricing_input_per_m": 15, "pricing_output_per_m": 75, "openness": "Proprietary", "key_strength": "48h Freeze Window Active", "value_index": 0.9, "context_window": "1000K", "delta_prev": "NEW", "trend": "flat", "release_date": "2026-09-01", "benchmarks": { "gdpval": 28.5, "epoch_eci": 28.1, "gpqa_diamond": 66, "hle": 39.1, "frontier_math": 21.8, "arc_agi_3": 61, "swe_bench_verified": 62.2, "terminal_bench": 67.8, "live_code_bench": 61.5, "scicode": 59.1, "osworld": 52.2, "browse_comp": 59.1, "automation_bench": 61.5, "gaia": 58.4, "vectara_hallucination": 2.8, "facts_grounding": 92.8, "uptime_pct": 99.85, "pricing_input_per_m": 15, "pricing_output_per_m": 75, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1333, "openrouter_rank": 53 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-mythos-preview-hold", "official_docs_url": "https://aki1k.com/models/claude-mythos-preview-hold", "eval_audit_url": "https://artificialanalysis.ai/models/claude-mythos-preview-hold", "sha256": "sha256_2026_53_claude-mythos-pr" }, { "rank": 54, "id": "deepseek-r2", "name": "DeepSeek R2", "provider": "DeepSeek AI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 61.5, "aa_intelligence": 27.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.3, "pricing_output_per_m": 0.9, "openness": "Open Weights", "key_strength": "Open Reasoning Reinforcement Loop", "value_index": 73.8, "context_window": "128K", "delta_prev": "+0.3", "trend": "up", "release_date": "2026-02-25", "benchmarks": { "gdpval": 28, "epoch_eci": 27.6, "gpqa_diamond": 65.7, "hle": 38.8, "frontier_math": 21.7, "arc_agi_3": 60.7, "swe_bench_verified": 61.8, "terminal_bench": 67.4, "live_code_bench": 61.2, "scicode": 58.7, "osworld": 51.9, "browse_comp": 58.7, "automation_bench": 61.2, "gaia": 58.1, "vectara_hallucination": 2.8, "facts_grounding": 92.7, "uptime_pct": 99.85, "pricing_input_per_m": 0.3, "pricing_output_per_m": 0.9, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1332, "openrouter_rank": 54 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-r2", "official_docs_url": "https://aki1k.com/models/deepseek-r2", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-r2", "sha256": "sha256_2026_54_deepseek-r2" }, { "rank": 55, "id": "grok-3", "name": "Grok 3", "provider": "xAI", "tier": "Tier 3: Enterprise Reliable", "aki_score": 61, "aa_intelligence": 27, "confidence_interval_95": 0.6, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "openness": "Proprietary", "key_strength": "Colossus Early Foundation Run", "value_index": 22, "context_window": "512K", "delta_prev": "-0.3", "trend": "down", "release_date": "2025-12-15", "benchmarks": { "gdpval": 27.5, "epoch_eci": 27.1, "gpqa_diamond": 65.4, "hle": 38.5, "frontier_math": 21.5, "arc_agi_3": 60.4, "swe_bench_verified": 61.5, "terminal_bench": 67.1, "live_code_bench": 60.9, "scicode": 58.4, "osworld": 51.6, "browse_comp": 58.4, "automation_bench": 60.9, "gaia": 57.8, "vectara_hallucination": 2.9, "facts_grounding": 92.6, "uptime_pct": 99.85, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1331, "openrouter_rank": 55 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/grok-3", "official_docs_url": "https://aki1k.com/models/grok-3", "eval_audit_url": "https://artificialanalysis.ai/models/grok-3", "sha256": "sha256_2026_55_grok-3" }, { "rank": 56, "id": "gpt-6-luna-ultra-compact", "name": "GPT-6 Luna Ultra-Compact", "provider": "OpenAI", "tier": "Tier 4: Cost Optimized", "aki_score": 60.5, "aa_intelligence": 26.5, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Proprietary", "key_strength": "Distilled Edge Foundation", "value_index": 435.6, "context_window": "128K", "delta_prev": "+0.5", "trend": "up", "release_date": "2026-08-15", "benchmarks": { "gdpval": 27, "epoch_eci": 26.6, "gpqa_diamond": 65.1, "hle": 38.3, "frontier_math": 21.4, "arc_agi_3": 60.1, "swe_bench_verified": 61.1, "terminal_bench": 66.7, "live_code_bench": 60.5, "scicode": 58.1, "osworld": 51.3, "browse_comp": 58.1, "automation_bench": 60.5, "gaia": 57.5, "vectara_hallucination": 2.9, "facts_grounding": 92.6, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1330, "openrouter_rank": 56 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-6-luna-ultra-compact", "official_docs_url": "https://aki1k.com/models/gpt-6-luna-ultra-compact", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-6-luna-ultra-compact", "sha256": "sha256_2026_56_gpt-6-luna-ultra" }, { "rank": 57, "id": "claude-3-7-sonnet-historical", "name": "Claude 3.7 Sonnet [Historical]", "provider": "Anthropic", "tier": "Tier 4: Cost Optimized", "aki_score": 60, "aa_intelligence": 26, "confidence_interval_95": 0.3, "pricing_input_per_m": 3, "pricing_output_per_m": 15, "openness": "Proprietary", "key_strength": "Historical 2024/2025 Frontier Baseline", "value_index": 4.5, "context_window": "200K", "delta_prev": "-1.2", "trend": "down", "release_date": "2025-02-24", "benchmarks": { "gdpval": 26.5, "epoch_eci": 26.1, "gpqa_diamond": 64.8, "hle": 38, "frontier_math": 21.2, "arc_agi_3": 59.8, "swe_bench_verified": 60.8, "terminal_bench": 66.4, "live_code_bench": 60.2, "scicode": 57.8, "osworld": 51, "browse_comp": 57.8, "automation_bench": 60.2, "gaia": 57.2, "vectara_hallucination": 2.9, "facts_grounding": 92.5, "uptime_pct": 99.85, "pricing_input_per_m": 3, "pricing_output_per_m": 15, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1329, "openrouter_rank": 57 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-3-7-sonnet-historical", "official_docs_url": "https://aki1k.com/models/claude-3-7-sonnet-historical", "eval_audit_url": "https://artificialanalysis.ai/models/claude-3-7-sonnet-historical", "sha256": "sha256_2026_57_claude-3-7-sonne" }, { "rank": 58, "id": "openai-o3-historical", "name": "OpenAI o3 [Historical]", "provider": "OpenAI", "tier": "Tier 4: Cost Optimized", "aki_score": 59.6, "aa_intelligence": 25.6, "confidence_interval_95": 0.4, "pricing_input_per_m": 10, "pricing_output_per_m": 40, "openness": "Proprietary", "key_strength": "Historical Olympiad Benchmark Baseline", "value_index": 1.7, "context_window": "200K", "delta_prev": "-1.4", "trend": "down", "release_date": "2025-01-31", "benchmarks": { "gdpval": 26.1, "epoch_eci": 25.7, "gpqa_diamond": 64.6, "hle": 37.8, "frontier_math": 21.1, "arc_agi_3": 59.6, "swe_bench_verified": 60.5, "terminal_bench": 66.1, "live_code_bench": 59.9, "scicode": 57.5, "osworld": 50.8, "browse_comp": 57.5, "automation_bench": 59.9, "gaia": 57, "vectara_hallucination": 2.9, "facts_grounding": 92.5, "uptime_pct": 99.85, "pricing_input_per_m": 10, "pricing_output_per_m": 40, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1328, "openrouter_rank": 58 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/openai-o3-historical", "official_docs_url": "https://aki1k.com/models/openai-o3-historical", "eval_audit_url": "https://artificialanalysis.ai/models/openai-o3-historical", "sha256": "sha256_2026_58_openai-o3-histor" }, { "rank": 59, "id": "qwen-2-5-max", "name": "Qwen 2.5 Max", "provider": "Alibaba Cloud", "tier": "Tier 4: Cost Optimized", "aki_score": 59.2, "aa_intelligence": 25.1, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "openness": "Open Weights", "key_strength": "2024 Chinese Frontier Baseline", "value_index": 26.6, "context_window": "128K", "delta_prev": "-0.8", "trend": "down", "release_date": "2024-11-15", "benchmarks": { "gdpval": 25.6, "epoch_eci": 25.2, "gpqa_diamond": 64.3, "hle": 37.6, "frontier_math": 20.9, "arc_agi_3": 59.3, "swe_bench_verified": 60.3, "terminal_bench": 65.8, "live_code_bench": 59.7, "scicode": 57.3, "osworld": 50.5, "browse_comp": 57.3, "automation_bench": 59.7, "gaia": 56.7, "vectara_hallucination": 2.9, "facts_grounding": 92.4, "uptime_pct": 99.85, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1327, "openrouter_rank": 59 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/qwen-2-5-max", "official_docs_url": "https://aki1k.com/models/qwen-2-5-max", "eval_audit_url": "https://artificialanalysis.ai/models/qwen-2-5-max", "sha256": "sha256_2026_59_qwen-2-5-max" }, { "rank": 60, "id": "deepseek-v3", "name": "DeepSeek V3", "provider": "DeepSeek AI", "tier": "Tier 4: Cost Optimized", "aki_score": 58.8, "aa_intelligence": 24.7, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.14, "pricing_output_per_m": 0.28, "openness": "Open Weights", "key_strength": "2024 MLA MoE Breakthrough Baseline", "value_index": 216, "context_window": "128K", "delta_prev": "-0.6", "trend": "down", "release_date": "2024-12-26", "benchmarks": { "gdpval": 25.2, "epoch_eci": 24.8, "gpqa_diamond": 64.1, "hle": 37.3, "frontier_math": 20.8, "arc_agi_3": 59.1, "swe_bench_verified": 60, "terminal_bench": 65.6, "live_code_bench": 59.4, "scicode": 57, "osworld": 50.3, "browse_comp": 57, "automation_bench": 59.4, "gaia": 56.5, "vectara_hallucination": 2.9, "facts_grounding": 92.3, "uptime_pct": 99.85, "pricing_input_per_m": 0.14, "pricing_output_per_m": 0.28, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1326, "openrouter_rank": 60 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-v3", "official_docs_url": "https://aki1k.com/models/deepseek-v3", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-v3", "sha256": "sha256_2026_60_deepseek-v3" }, { "rank": 61, "id": "gemini-2-0-flash", "name": "Gemini 2.0 Flash", "provider": "Google DeepMind", "tier": "Tier 4: Cost Optimized", "aki_score": 58.4, "aa_intelligence": 24.2, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.4, "openness": "Proprietary", "key_strength": "2024 Fast Multimodal Baseline", "value_index": 161.7, "context_window": "1000K", "delta_prev": "-0.5", "trend": "down", "release_date": "2025-02-05", "benchmarks": { "gdpval": 24.7, "epoch_eci": 24.3, "gpqa_diamond": 63.9, "hle": 37.1, "frontier_math": 20.7, "arc_agi_3": 58.9, "swe_bench_verified": 59.7, "terminal_bench": 65.3, "live_code_bench": 59.1, "scicode": 56.8, "osworld": 50, "browse_comp": 56.8, "automation_bench": 59.1, "gaia": 56.2, "vectara_hallucination": 3, "facts_grounding": 92.3, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.4, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1326, "openrouter_rank": 61 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemini-2-0-flash", "official_docs_url": "https://aki1k.com/models/gemini-2-0-flash", "eval_audit_url": "https://artificialanalysis.ai/models/gemini-2-0-flash", "sha256": "sha256_2026_61_gemini-2-0-flash" }, { "rank": 62, "id": "claude-3-5-sonnet-v2", "name": "Claude 3.5 Sonnet v2", "provider": "Anthropic", "tier": "Tier 4: Cost Optimized", "aki_score": 58, "aa_intelligence": 23.8, "confidence_interval_95": 0.3, "pricing_input_per_m": 3, "pricing_output_per_m": 15, "openness": "Proprietary", "key_strength": "2024 SWE-bench Verified Pioneer", "value_index": 4.3, "context_window": "200K", "delta_prev": "-1.0", "trend": "down", "release_date": "2024-10-22", "benchmarks": { "gdpval": 24.3, "epoch_eci": 23.9, "gpqa_diamond": 63.6, "hle": 36.9, "frontier_math": 20.6, "arc_agi_3": 58.6, "swe_bench_verified": 59.4, "terminal_bench": 65, "live_code_bench": 58.9, "scicode": 56.5, "osworld": 49.8, "browse_comp": 56.5, "automation_bench": 58.9, "gaia": 56, "vectara_hallucination": 3, "facts_grounding": 92.2, "uptime_pct": 99.85, "pricing_input_per_m": 3, "pricing_output_per_m": 15, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1325, "openrouter_rank": 62 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-3-5-sonnet-v2", "official_docs_url": "https://aki1k.com/models/claude-3-5-sonnet-v2", "eval_audit_url": "https://artificialanalysis.ai/models/claude-3-5-sonnet-v2", "sha256": "sha256_2026_62_claude-3-5-sonne" }, { "rank": 63, "id": "openai-o1", "name": "OpenAI o1", "provider": "OpenAI", "tier": "Tier 4: Cost Optimized", "aki_score": 57.6, "aa_intelligence": 23.3, "confidence_interval_95": 0.4, "pricing_input_per_m": 15, "pricing_output_per_m": 60, "openness": "Proprietary", "key_strength": "2024 First Reasoner Baseline", "value_index": 1.1, "context_window": "128K", "delta_prev": "-1.5", "trend": "down", "release_date": "2024-09-12", "benchmarks": { "gdpval": 23.8, "epoch_eci": 23.4, "gpqa_diamond": 63.4, "hle": 36.7, "frontier_math": 20.4, "arc_agi_3": 58.4, "swe_bench_verified": 59.2, "terminal_bench": 64.7, "live_code_bench": 58.6, "scicode": 56.3, "osworld": 49.6, "browse_comp": 56.3, "automation_bench": 58.6, "gaia": 55.7, "vectara_hallucination": 3, "facts_grounding": 92.2, "uptime_pct": 99.85, "pricing_input_per_m": 15, "pricing_output_per_m": 60, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1324, "openrouter_rank": 63 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/openai-o1", "official_docs_url": "https://aki1k.com/models/openai-o1", "eval_audit_url": "https://artificialanalysis.ai/models/openai-o1", "sha256": "sha256_2026_63_openai-o1" }, { "rank": 64, "id": "gpt-4o", "name": "GPT-4o", "provider": "OpenAI", "tier": "Tier 4: Cost Optimized", "aki_score": 57.2, "aa_intelligence": 22.9, "confidence_interval_95": 0.3, "pricing_input_per_m": 2.5, "pricing_output_per_m": 10, "openness": "Proprietary", "key_strength": "2024 Omni Multimodal Baseline", "value_index": 6.3, "context_window": "128K", "delta_prev": "-1.2", "trend": "down", "release_date": "2024-05-13", "benchmarks": { "gdpval": 23.4, "epoch_eci": 23, "gpqa_diamond": 63.2, "hle": 36.5, "frontier_math": 20.3, "arc_agi_3": 58.2, "swe_bench_verified": 58.9, "terminal_bench": 64.5, "live_code_bench": 58.3, "scicode": 56, "osworld": 49.3, "browse_comp": 56, "automation_bench": 58.3, "gaia": 55.5, "vectara_hallucination": 3, "facts_grounding": 92.2, "uptime_pct": 99.85, "pricing_input_per_m": 2.5, "pricing_output_per_m": 10, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1323, "openrouter_rank": 64 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-4o", "official_docs_url": "https://aki1k.com/models/gpt-4o", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-4o", "sha256": "sha256_2026_64_gpt-4o" }, { "rank": 65, "id": "claude-3-5-haiku", "name": "Claude 3.5 Haiku", "provider": "Anthropic", "tier": "Tier 4: Cost Optimized", "aki_score": 56.8, "aa_intelligence": 22.4, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.8, "pricing_output_per_m": 4, "openness": "Proprietary", "key_strength": "2024 Fast Classification Baseline", "value_index": 16, "context_window": "200K", "delta_prev": "-0.7", "trend": "down", "release_date": "2024-11-04", "benchmarks": { "gdpval": 22.9, "epoch_eci": 22.5, "gpqa_diamond": 62.9, "hle": 36.2, "frontier_math": 20.2, "arc_agi_3": 57.9, "swe_bench_verified": 58.6, "terminal_bench": 64.2, "live_code_bench": 58.1, "scicode": 55.8, "osworld": 49.1, "browse_comp": 55.8, "automation_bench": 58.1, "gaia": 55.2, "vectara_hallucination": 3, "facts_grounding": 92.1, "uptime_pct": 99.85, "pricing_input_per_m": 0.8, "pricing_output_per_m": 4, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1322, "openrouter_rank": 65 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-3-5-haiku", "official_docs_url": "https://aki1k.com/models/claude-3-5-haiku", "eval_audit_url": "https://artificialanalysis.ai/models/claude-3-5-haiku", "sha256": "sha256_2026_65_claude-3-5-haiku" }, { "rank": 66, "id": "llama-3-3-70b", "name": "Llama 3.3 70B", "provider": "Meta AI", "tier": "Tier 4: Cost Optimized", "aki_score": 56.4, "aa_intelligence": 22, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.35, "pricing_output_per_m": 0.7, "openness": "Open Weights", "key_strength": "2024 Open Weights Workhorse", "value_index": 82.9, "context_window": "128K", "delta_prev": "-0.5", "trend": "down", "release_date": "2024-12-06", "benchmarks": { "gdpval": 22.5, "epoch_eci": 22.1, "gpqa_diamond": 62.7, "hle": 36, "frontier_math": 20, "arc_agi_3": 57.7, "swe_bench_verified": 58.4, "terminal_bench": 63.9, "live_code_bench": 57.8, "scicode": 55.5, "osworld": 48.8, "browse_comp": 55.5, "automation_bench": 57.8, "gaia": 55, "vectara_hallucination": 3, "facts_grounding": 92, "uptime_pct": 99.85, "pricing_input_per_m": 0.35, "pricing_output_per_m": 0.7, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1321, "openrouter_rank": 66 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-3-3-70b", "official_docs_url": "https://aki1k.com/models/llama-3-3-70b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-3-3-70b", "sha256": "sha256_2026_66_llama-3-3-70b" }, { "rank": 67, "id": "mistral-nemo-22b", "name": "Mistral Nemo 22B", "provider": "Mistral AI", "tier": "Tier 4: Cost Optimized", "aki_score": 56, "aa_intelligence": 21.6, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "openness": "Open Weights", "key_strength": "2024 Small Model Benchmark", "value_index": 201.6, "context_window": "128K", "delta_prev": "-0.4", "trend": "down", "release_date": "2024-07-18", "benchmarks": { "gdpval": 22.1, "epoch_eci": 21.7, "gpqa_diamond": 62.5, "hle": 35.8, "frontier_math": 19.9, "arc_agi_3": 57.5, "swe_bench_verified": 58.1, "terminal_bench": 63.6, "live_code_bench": 57.5, "scicode": 55.3, "osworld": 48.6, "browse_comp": 55.3, "automation_bench": 57.5, "gaia": 54.7, "vectara_hallucination": 3, "facts_grounding": 92, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "tokens_per_sec": 80, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1320, "openrouter_rank": 67 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/mistral-nemo-22b", "official_docs_url": "https://aki1k.com/models/mistral-nemo-22b", "eval_audit_url": "https://artificialanalysis.ai/models/mistral-nemo-22b", "sha256": "sha256_2026_67_mistral-nemo-22b" }, { "rank": 68, "id": "nova-lite", "name": "Nova Lite", "provider": "Amazon AWS", "tier": "Tier 4: Cost Optimized", "aki_score": 55.6, "aa_intelligence": 21.1, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.6, "openness": "Proprietary", "key_strength": "2024 Low-Cost AWS Baseline", "value_index": 102.6, "context_window": "128K", "delta_prev": "-0.3", "trend": "down", "release_date": "2024-12-03", "benchmarks": { "gdpval": 21.6, "epoch_eci": 21.2, "gpqa_diamond": 62.2, "hle": 35.6, "frontier_math": 19.8, "arc_agi_3": 57.2, "swe_bench_verified": 57.8, "terminal_bench": 63.4, "live_code_bench": 57.3, "scicode": 55, "osworld": 48.4, "browse_comp": 55, "automation_bench": 57.3, "gaia": 54.5, "vectara_hallucination": 3.1, "facts_grounding": 92, "uptime_pct": 99.85, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.6, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1320, "openrouter_rank": 68 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/nova-lite", "official_docs_url": "https://aki1k.com/models/nova-lite", "eval_audit_url": "https://artificialanalysis.ai/models/nova-lite", "sha256": "sha256_2026_68_nova-lite" }, { "rank": 69, "id": "command-r-08-24", "name": "Command R 08-24", "provider": "Cohere", "tier": "Tier 4: Cost Optimized", "aki_score": 55.2, "aa_intelligence": 20.7, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "openness": "Proprietary", "key_strength": "2024 Enterprise RAG Baseline", "value_index": 39.7, "context_window": "128K", "delta_prev": "-0.4", "trend": "down", "release_date": "2024-08-20", "benchmarks": { "gdpval": 21.2, "epoch_eci": 20.8, "gpqa_diamond": 62, "hle": 35.4, "frontier_math": 19.7, "arc_agi_3": 57, "swe_bench_verified": 57.5, "terminal_bench": 63.1, "live_code_bench": 57, "scicode": 54.8, "osworld": 48.1, "browse_comp": 54.8, "automation_bench": 57, "gaia": 54.2, "vectara_hallucination": 3.1, "facts_grounding": 91.9, "uptime_pct": 99.85, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1319, "openrouter_rank": 69 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/command-r-08-24", "official_docs_url": "https://aki1k.com/models/command-r-08-24", "eval_audit_url": "https://artificialanalysis.ai/models/command-r-08-24", "sha256": "sha256_2026_69_command-r-08-24" }, { "rank": 70, "id": "starcoder2-15b", "name": "StarCoder2 15B", "provider": "BigCode", "tier": "Tier 4: Cost Optimized", "aki_score": 54.8, "aa_intelligence": 20.2, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Apache 2.0", "key_strength": "2024 Open Code Model Baseline", "value_index": 394.6, "context_window": "64K", "delta_prev": "-0.2", "trend": "flat", "release_date": "2024-02-28", "benchmarks": { "gdpval": 20.7, "epoch_eci": 20.3, "gpqa_diamond": 61.8, "hle": 35.1, "frontier_math": 19.5, "arc_agi_3": 56.8, "swe_bench_verified": 57.3, "terminal_bench": 62.8, "live_code_bench": 56.7, "scicode": 54.5, "osworld": 47.9, "browse_comp": 54.5, "automation_bench": 56.7, "gaia": 54, "vectara_hallucination": 3.1, "facts_grounding": 91.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 40, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1318, "openrouter_rank": 70 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/starcoder2-15b", "official_docs_url": "https://aki1k.com/models/starcoder2-15b", "eval_audit_url": "https://artificialanalysis.ai/models/starcoder2-15b", "sha256": "sha256_2026_70_starcoder2-15b" }, { "rank": 71, "id": "codellama-70b", "name": "CodeLlama 70B", "provider": "Meta AI", "tier": "Tier 4: Cost Optimized", "aki_score": 54.4, "aa_intelligence": 19.8, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.5, "openness": "Open Weights", "key_strength": "2023/2024 Code Generation Baseline", "value_index": 115.2, "context_window": "64K", "delta_prev": "-0.3", "trend": "down", "release_date": "2024-01-29", "benchmarks": { "gdpval": 20.3, "epoch_eci": 19.9, "gpqa_diamond": 61.6, "hle": 34.9, "frontier_math": 19.4, "arc_agi_3": 56.6, "swe_bench_verified": 57, "terminal_bench": 62.5, "live_code_bench": 56.4, "scicode": 54.3, "osworld": 47.6, "browse_comp": 54.3, "automation_bench": 56.4, "gaia": 53.7, "vectara_hallucination": 3.1, "facts_grounding": 91.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.5, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1317, "openrouter_rank": 71 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/codellama-70b", "official_docs_url": "https://aki1k.com/models/codellama-70b", "eval_audit_url": "https://artificialanalysis.ai/models/codellama-70b", "sha256": "sha256_2026_71_codellama-70b" }, { "rank": 72, "id": "phi-4-5-moe", "name": "Phi-4.5 MoE", "provider": "Microsoft", "tier": "Tier 4: Cost Optimized", "aki_score": 54, "aa_intelligence": 19.3, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "openness": "MIT", "key_strength": "2024 Edge Reasoning MoE", "value_index": 194.4, "context_window": "64K", "delta_prev": "+0.1", "trend": "flat", "release_date": "2024-12-12", "benchmarks": { "gdpval": 19.8, "epoch_eci": 19.4, "gpqa_diamond": 61.3, "hle": 34.7, "frontier_math": 19.3, "arc_agi_3": 56.3, "swe_bench_verified": 56.7, "terminal_bench": 62.3, "live_code_bench": 56.2, "scicode": 54, "osworld": 47.4, "browse_comp": 54, "automation_bench": 56.2, "gaia": 53.5, "vectara_hallucination": 3.1, "facts_grounding": 91.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "tokens_per_sec": 80, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1316, "openrouter_rank": 72 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/phi-4-5-moe", "official_docs_url": "https://aki1k.com/models/phi-4-5-moe", "eval_audit_url": "https://artificialanalysis.ai/models/phi-4-5-moe", "sha256": "sha256_2026_72_phi-4-5-moe" }, { "rank": 73, "id": "gemma-2-27b", "name": "Gemma 2 27B", "provider": "Google DeepMind", "tier": "Tier 4: Cost Optimized", "aki_score": 53.6, "aa_intelligence": 18.9, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Open Weights", "key_strength": "2024 Open Architecture Baseline", "value_index": 128.6, "context_window": "64K", "delta_prev": "-0.3", "trend": "down", "release_date": "2024-06-27", "benchmarks": { "gdpval": 19.4, "epoch_eci": 19, "gpqa_diamond": 61.1, "hle": 34.5, "frontier_math": 19.2, "arc_agi_3": 56.1, "swe_bench_verified": 56.4, "terminal_bench": 62, "live_code_bench": 55.9, "scicode": 53.8, "osworld": 47.2, "browse_comp": 53.8, "automation_bench": 55.9, "gaia": 53.2, "vectara_hallucination": 3.1, "facts_grounding": 91.7, "uptime_pct": 99.85, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 100, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1315, "openrouter_rank": 73 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemma-2-27b", "official_docs_url": "https://aki1k.com/models/gemma-2-27b", "eval_audit_url": "https://artificialanalysis.ai/models/gemma-2-27b", "sha256": "sha256_2026_73_gemma-2-27b" }, { "rank": 74, "id": "gemma-2-9b", "name": "Gemma 2 9B", "provider": "Google DeepMind", "tier": "Tier 4: Cost Optimized", "aki_score": 53.2, "aa_intelligence": 18.4, "confidence_interval_95": 0.3, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Open Weights", "key_strength": "2024 Lightweight Open Baseline", "value_index": 383, "context_window": "32K", "delta_prev": "-0.2", "trend": "down", "release_date": "2024-06-27", "benchmarks": { "gdpval": 18.9, "epoch_eci": 18.5, "gpqa_diamond": 60.9, "hle": 34.3, "frontier_math": 19, "arc_agi_3": 55.9, "swe_bench_verified": 56.2, "terminal_bench": 61.7, "live_code_bench": 55.6, "scicode": 53.5, "osworld": 46.9, "browse_comp": 53.5, "automation_bench": 55.6, "gaia": 53, "vectara_hallucination": 3.1, "facts_grounding": 91.7, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1314, "openrouter_rank": 74 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gemma-2-9b", "official_docs_url": "https://aki1k.com/models/gemma-2-9b", "eval_audit_url": "https://artificialanalysis.ai/models/gemma-2-9b", "sha256": "sha256_2026_74_gemma-2-9b" }, { "rank": 75, "id": "deepseek-coder-33b", "name": "DeepSeek Coder 33B", "provider": "DeepSeek AI", "tier": "Tier 4: Cost Optimized", "aki_score": 52.8, "aa_intelligence": 18, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.2, "openness": "Open Weights", "key_strength": "2023/2024 Coding Predecessor", "value_index": 271.5, "context_window": "64K", "delta_prev": "-0.4", "trend": "down", "release_date": "2023-11-20", "benchmarks": { "gdpval": 18.5, "epoch_eci": 18.1, "gpqa_diamond": 60.6, "hle": 34, "frontier_math": 18.9, "arc_agi_3": 55.6, "swe_bench_verified": 55.9, "terminal_bench": 61.4, "live_code_bench": 55.4, "scicode": 53.3, "osworld": 46.7, "browse_comp": 53.3, "automation_bench": 55.4, "gaia": 52.7, "vectara_hallucination": 3.2, "facts_grounding": 91.6, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.2, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1314, "openrouter_rank": 75 }, "replication_count": 4, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/deepseek-coder-33b", "official_docs_url": "https://aki1k.com/models/deepseek-coder-33b", "eval_audit_url": "https://artificialanalysis.ai/models/deepseek-coder-33b", "sha256": "sha256_2026_75_deepseek-coder-3" }, { "rank": 76, "id": "falcon-180b", "name": "Falcon 180B", "provider": "TII", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 52.1, "aa_intelligence": 17.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.2, "openness": "Apache 2.0", "key_strength": "Early UAE Open Frontier Milestone", "value_index": 46.9, "context_window": "32K", "delta_prev": "-0.5", "trend": "down", "release_date": "2023-09-06", "benchmarks": { "gdpval": 18, "epoch_eci": 17.6, "gpqa_diamond": 60.2, "hle": 33.7, "frontier_math": 18.7, "arc_agi_3": 55.2, "swe_bench_verified": 55.4, "terminal_bench": 60.9, "live_code_bench": 54.9, "scicode": 52.8, "osworld": 46.3, "browse_comp": 52.8, "automation_bench": 54.9, "gaia": 52.3, "vectara_hallucination": 3.2, "facts_grounding": 91.5, "uptime_pct": 99.85, "pricing_input_per_m": 0.4, "pricing_output_per_m": 1.2, "tokens_per_sec": 60, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1312, "openrouter_rank": 76 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/falcon-180b", "official_docs_url": "https://aki1k.com/models/falcon-180b", "eval_audit_url": "https://artificialanalysis.ai/models/falcon-180b", "sha256": "sha256_2026_76_falcon-180b" }, { "rank": 77, "id": "mistral-7b-v0-4", "name": "Mistral 7B v0.4", "provider": "Mistral AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 51.5, "aa_intelligence": 17, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Apache 2.0", "key_strength": "Historical Efficient Dense Foundation", "value_index": 370.8, "context_window": "32K", "delta_prev": "-0.4", "trend": "down", "release_date": "2023-09-27", "benchmarks": { "gdpval": 17.5, "epoch_eci": 17.1, "gpqa_diamond": 59.9, "hle": 33.3, "frontier_math": 18.5, "arc_agi_3": 54.9, "swe_bench_verified": 55, "terminal_bench": 60.5, "live_code_bench": 54.5, "scicode": 52.4, "osworld": 45.9, "browse_comp": 52.4, "automation_bench": 54.5, "gaia": 51.9, "vectara_hallucination": 3.2, "facts_grounding": 91.4, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 80, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1311, "openrouter_rank": 77 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/mistral-7b-v0-4", "official_docs_url": "https://aki1k.com/models/mistral-7b-v0-4", "eval_audit_url": "https://artificialanalysis.ai/models/mistral-7b-v0-4", "sha256": "sha256_2026_77_mistral-7b-v0-4" }, { "rank": 78, "id": "llama-3-1-405b", "name": "Llama 3.1 405B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 51, "aa_intelligence": 16.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "openness": "Open Weights", "key_strength": "2024 Flagship Open Architecture", "value_index": 18.4, "context_window": "128K", "delta_prev": "-1.1", "trend": "down", "release_date": "2024-07-23", "benchmarks": { "gdpval": 17, "epoch_eci": 16.6, "gpqa_diamond": 59.6, "hle": 33, "frontier_math": 18.3, "arc_agi_3": 54.6, "swe_bench_verified": 54.7, "terminal_bench": 60.2, "live_code_bench": 54.2, "scicode": 52.1, "osworld": 45.6, "browse_comp": 52.1, "automation_bench": 54.2, "gaia": 51.6, "vectara_hallucination": 3.2, "facts_grounding": 91.4, "uptime_pct": 99.85, "pricing_input_per_m": 1, "pricing_output_per_m": 3, "tokens_per_sec": 100, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1310, "openrouter_rank": 78 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-3-1-405b", "official_docs_url": "https://aki1k.com/models/llama-3-1-405b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-3-1-405b", "sha256": "sha256_2026_78_llama-3-1-405b" }, { "rank": 79, "id": "llama-3-1-70b", "name": "Llama 3.1 70B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 50.4, "aa_intelligence": 16, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.3, "pricing_output_per_m": 0.6, "openness": "Open Weights", "key_strength": "2024 Enterprise Open Baseline", "value_index": 86.4, "context_window": "128K", "delta_prev": "-0.8", "trend": "down", "release_date": "2024-07-23", "benchmarks": { "gdpval": 16.5, "epoch_eci": 16.1, "gpqa_diamond": 59.2, "hle": 32.7, "frontier_math": 18.1, "arc_agi_3": 54.2, "swe_bench_verified": 54.3, "terminal_bench": 59.8, "live_code_bench": 53.8, "scicode": 51.8, "osworld": 45.2, "browse_comp": 51.8, "automation_bench": 53.8, "gaia": 51.2, "vectara_hallucination": 3.2, "facts_grounding": 91.3, "uptime_pct": 99.85, "pricing_input_per_m": 0.3, "pricing_output_per_m": 0.6, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1308, "openrouter_rank": 79 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-3-1-70b", "official_docs_url": "https://aki1k.com/models/llama-3-1-70b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-3-1-70b", "sha256": "sha256_2026_79_llama-3-1-70b" }, { "rank": 80, "id": "llama-2-70b", "name": "Llama 2 70B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 49.8, "aa_intelligence": 15.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.4, "pricing_output_per_m": 0.8, "openness": "Open Weights", "key_strength": "2023 Open Source Watershed Baseline", "value_index": 64, "context_window": "4k", "delta_prev": "-0.5", "trend": "down", "release_date": "2023-07-18", "benchmarks": { "gdpval": 16, "epoch_eci": 15.6, "gpqa_diamond": 58.9, "hle": 32.4, "frontier_math": 17.9, "arc_agi_3": 53.9, "swe_bench_verified": 53.9, "terminal_bench": 59.4, "live_code_bench": 53.4, "scicode": 51.4, "osworld": 44.9, "browse_comp": 51.4, "automation_bench": 53.4, "gaia": 50.9, "vectara_hallucination": 3.3, "facts_grounding": 91.2, "uptime_pct": 99.85, "pricing_input_per_m": 0.4, "pricing_output_per_m": 0.8, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1307, "openrouter_rank": 80 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-2-70b", "official_docs_url": "https://aki1k.com/models/llama-2-70b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-2-70b", "sha256": "sha256_2026_80_llama-2-70b" }, { "rank": 81, "id": "claude-3-opus", "name": "Claude 3 Opus", "provider": "Anthropic", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 49.2, "aa_intelligence": 15, "confidence_interval_95": 0.4, "pricing_input_per_m": 15, "pricing_output_per_m": 75, "openness": "Proprietary", "key_strength": "2024 Q1 Reasoning Vanguard Baseline", "value_index": 0.7, "context_window": "200K", "delta_prev": "-1.5", "trend": "down", "release_date": "2024-03-04", "benchmarks": { "gdpval": 15.5, "epoch_eci": 15.1, "gpqa_diamond": 58.5, "hle": 32.1, "frontier_math": 17.7, "arc_agi_3": 53.5, "swe_bench_verified": 53.5, "terminal_bench": 58.9, "live_code_bench": 53, "scicode": 51, "osworld": 44.5, "browse_comp": 51, "automation_bench": 53, "gaia": 50.5, "vectara_hallucination": 3.3, "facts_grounding": 91.2, "uptime_pct": 99.85, "pricing_input_per_m": 15, "pricing_output_per_m": 75, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1306, "openrouter_rank": 81 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-3-opus", "official_docs_url": "https://aki1k.com/models/claude-3-opus", "eval_audit_url": "https://artificialanalysis.ai/models/claude-3-opus", "sha256": "sha256_2026_81_claude-3-opus" }, { "rank": 82, "id": "gpt-4-turbo", "name": "GPT-4 Turbo", "provider": "OpenAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 48.6, "aa_intelligence": 14.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 10, "pricing_output_per_m": 30, "openness": "Proprietary", "key_strength": "2023 Developer Conference Pioneer", "value_index": 1.7, "context_window": "128K", "delta_prev": "-1.2", "trend": "down", "release_date": "2023-11-06", "benchmarks": { "gdpval": 15, "epoch_eci": 14.6, "gpqa_diamond": 58.2, "hle": 31.7, "frontier_math": 17.6, "arc_agi_3": 53.2, "swe_bench_verified": 53, "terminal_bench": 58.5, "live_code_bench": 52.6, "scicode": 50.6, "osworld": 44.2, "browse_comp": 50.6, "automation_bench": 52.6, "gaia": 50.1, "vectara_hallucination": 3.3, "facts_grounding": 91.1, "uptime_pct": 99.85, "pricing_input_per_m": 10, "pricing_output_per_m": 30, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1304, "openrouter_rank": 82 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-4-turbo", "official_docs_url": "https://aki1k.com/models/gpt-4-turbo", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-4-turbo", "sha256": "sha256_2026_82_gpt-4-turbo" }, { "rank": 83, "id": "palm-2-l", "name": "PaLM 2 L", "provider": "Google", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 48, "aa_intelligence": 14, "confidence_interval_95": 0.5, "pricing_input_per_m": 1.5, "pricing_output_per_m": 4.5, "openness": "Proprietary", "key_strength": "2023 Google I/O Foundation Baseline", "value_index": 11.5, "context_window": "8k", "delta_prev": "-0.6", "trend": "down", "release_date": "2023-05-10", "benchmarks": { "gdpval": 14.5, "epoch_eci": 14.1, "gpqa_diamond": 57.8, "hle": 31.4, "frontier_math": 17.4, "arc_agi_3": 52.8, "swe_bench_verified": 52.6, "terminal_bench": 58.1, "live_code_bench": 52.2, "scicode": 50.2, "osworld": 43.8, "browse_comp": 50.2, "automation_bench": 52.2, "gaia": 49.8, "vectara_hallucination": 3.3, "facts_grounding": 91, "uptime_pct": 99.85, "pricing_input_per_m": 1.5, "pricing_output_per_m": 4.5, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1303, "openrouter_rank": 83 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/palm-2-l", "official_docs_url": "https://aki1k.com/models/palm-2-l", "eval_audit_url": "https://artificialanalysis.ai/models/palm-2-l", "sha256": "sha256_2026_83_palm-2-l" }, { "rank": 84, "id": "claude-instant", "name": "Claude Instant", "provider": "Anthropic", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 47.4, "aa_intelligence": 13.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "openness": "Proprietary", "key_strength": "2023 Low-Latency Baseline", "value_index": 21.3, "context_window": "100K", "delta_prev": "-0.4", "trend": "down", "release_date": "2023-08-01", "benchmarks": { "gdpval": 14, "epoch_eci": 13.6, "gpqa_diamond": 57.5, "hle": 31.1, "frontier_math": 17.2, "arc_agi_3": 52.5, "swe_bench_verified": 52.2, "terminal_bench": 57.7, "live_code_bench": 51.8, "scicode": 49.9, "osworld": 43.4, "browse_comp": 49.9, "automation_bench": 51.8, "gaia": 49.4, "vectara_hallucination": 3.3, "facts_grounding": 90.9, "uptime_pct": 99.85, "pricing_input_per_m": 0.8, "pricing_output_per_m": 2.4, "tokens_per_sec": 120, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1302, "openrouter_rank": 84 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/claude-instant", "official_docs_url": "https://aki1k.com/models/claude-instant", "eval_audit_url": "https://artificialanalysis.ai/models/claude-instant", "sha256": "sha256_2026_84_claude-instant" }, { "rank": 85, "id": "cohere-command", "name": "Cohere Command", "provider": "Cohere", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 46.8, "aa_intelligence": 13, "confidence_interval_95": 0.5, "pricing_input_per_m": 1, "pricing_output_per_m": 2, "openness": "Proprietary", "key_strength": "2023 Early Enterprise Baseline", "value_index": 24.1, "context_window": "4k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-06-01", "benchmarks": { "gdpval": 13.5, "epoch_eci": 13.1, "gpqa_diamond": 57.1, "hle": 30.7, "frontier_math": 17, "arc_agi_3": 52.1, "swe_bench_verified": 51.8, "terminal_bench": 57.3, "live_code_bench": 51.4, "scicode": 49.5, "osworld": 43.1, "browse_comp": 49.5, "automation_bench": 51.4, "gaia": 49, "vectara_hallucination": 3.4, "facts_grounding": 90.8, "uptime_pct": 99.85, "pricing_input_per_m": 1, "pricing_output_per_m": 2, "tokens_per_sec": 40, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1301, "openrouter_rank": 85 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/cohere-command", "official_docs_url": "https://aki1k.com/models/cohere-command", "eval_audit_url": "https://artificialanalysis.ai/models/cohere-command", "sha256": "sha256_2026_85_cohere-command" }, { "rank": 86, "id": "falcon-40b", "name": "Falcon 40B", "provider": "TII", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 46.2, "aa_intelligence": 12.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.6, "openness": "Apache 2.0", "key_strength": "2023 TII Architecture Predecessor", "value_index": 83.2, "context_window": "2k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-05-25", "benchmarks": { "gdpval": 13, "epoch_eci": 12.6, "gpqa_diamond": 56.8, "hle": 30.4, "frontier_math": 16.8, "arc_agi_3": 51.8, "swe_bench_verified": 51.4, "terminal_bench": 56.9, "live_code_bench": 51, "scicode": 49.1, "osworld": 42.7, "browse_comp": 49.1, "automation_bench": 51, "gaia": 48.6, "vectara_hallucination": 3.4, "facts_grounding": 90.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.6, "tokens_per_sec": 60, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1299, "openrouter_rank": 86 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/falcon-40b", "official_docs_url": "https://aki1k.com/models/falcon-40b", "eval_audit_url": "https://artificialanalysis.ai/models/falcon-40b", "sha256": "sha256_2026_86_falcon-40b" }, { "rank": 87, "id": "mpt-30b", "name": "MPT-30B", "provider": "MosaicML", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 45.6, "aa_intelligence": 12, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Apache 2.0", "key_strength": "2023 Open Commercial Compute Baseline", "value_index": 109.4, "context_window": "8k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-06-22", "benchmarks": { "gdpval": 12.5, "epoch_eci": 12.1, "gpqa_diamond": 56.4, "hle": 30.1, "frontier_math": 16.6, "arc_agi_3": 51.4, "swe_bench_verified": 51, "terminal_bench": 56.5, "live_code_bench": 50.6, "scicode": 48.7, "osworld": 42.4, "browse_comp": 48.7, "automation_bench": 50.6, "gaia": 48.3, "vectara_hallucination": 3.4, "facts_grounding": 90.7, "uptime_pct": 99.85, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 80, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1298, "openrouter_rank": 87 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/mpt-30b", "official_docs_url": "https://aki1k.com/models/mpt-30b", "eval_audit_url": "https://artificialanalysis.ai/models/mpt-30b", "sha256": "sha256_2026_87_mpt-30b" }, { "rank": 88, "id": "llama-2-13b", "name": "Llama 2 13B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 45, "aa_intelligence": 11.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.2, "openness": "Open Weights", "key_strength": "2023 Mid-Size Open Baseline", "value_index": 231.4, "context_window": "4k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-07-18", "benchmarks": { "gdpval": 12, "epoch_eci": 11.6, "gpqa_diamond": 56.1, "hle": 29.8, "frontier_math": 16.4, "arc_agi_3": 51.1, "swe_bench_verified": 50.6, "terminal_bench": 56, "live_code_bench": 50.2, "scicode": 48.4, "osworld": 42, "browse_comp": 48.4, "automation_bench": 50.2, "gaia": 47.9, "vectara_hallucination": 3.4, "facts_grounding": 90.6, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.2, "tokens_per_sec": 100, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1297, "openrouter_rank": 88 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-2-13b", "official_docs_url": "https://aki1k.com/models/llama-2-13b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-2-13b", "sha256": "sha256_2026_88_llama-2-13b" }, { "rank": 89, "id": "vicuna-33b", "name": "Vicuna 33B", "provider": "LMSYS", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 44.4, "aa_intelligence": 11, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "openness": "Open Weights", "key_strength": "2023 Arena Evaluation Pioneer", "value_index": 106.6, "context_window": "2k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-06-15", "benchmarks": { "gdpval": 11.5, "epoch_eci": 11.1, "gpqa_diamond": 55.8, "hle": 29.4, "frontier_math": 16.2, "arc_agi_3": 50.8, "swe_bench_verified": 50.2, "terminal_bench": 55.6, "live_code_bench": 49.7, "scicode": 48, "osworld": 41.6, "browse_comp": 48, "automation_bench": 49.7, "gaia": 47.5, "vectara_hallucination": 3.4, "facts_grounding": 90.5, "uptime_pct": 99.85, "pricing_input_per_m": 0.15, "pricing_output_per_m": 0.45, "tokens_per_sec": 120, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1295, "openrouter_rank": 89 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/vicuna-33b", "official_docs_url": "https://aki1k.com/models/vicuna-33b", "eval_audit_url": "https://artificialanalysis.ai/models/vicuna-33b", "sha256": "sha256_2026_89_vicuna-33b" }, { "rank": 90, "id": "alpaca-65b", "name": "Alpaca 65B", "provider": "Stanford", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 43.8, "aa_intelligence": 10.5, "confidence_interval_95": 0.6, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.6, "openness": "Open Weights", "key_strength": "2023 Synthetic Instruction Pioneer", "value_index": 78.8, "context_window": "2k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-03-13", "benchmarks": { "gdpval": 11, "epoch_eci": 10.6, "gpqa_diamond": 55.4, "hle": 29.1, "frontier_math": 16, "arc_agi_3": 50.4, "swe_bench_verified": 49.8, "terminal_bench": 55.2, "live_code_bench": 49.3, "scicode": 47.6, "osworld": 41.3, "browse_comp": 47.6, "automation_bench": 49.3, "gaia": 47.2, "vectara_hallucination": 3.5, "facts_grounding": 90.5, "uptime_pct": 99.85, "pricing_input_per_m": 0.2, "pricing_output_per_m": 0.6, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1294, "openrouter_rank": 90 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/alpaca-65b", "official_docs_url": "https://aki1k.com/models/alpaca-65b", "eval_audit_url": "https://artificialanalysis.ai/models/alpaca-65b", "sha256": "sha256_2026_90_alpaca-65b" }, { "rank": 91, "id": "gpt-3-5-turbo", "name": "GPT-3.5 Turbo", "provider": "OpenAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 43.2, "aa_intelligence": 10, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "openness": "Proprietary", "key_strength": "2022/2023 ChatGPT Consumer Launch Milestone", "value_index": 31.1, "context_window": "16K", "delta_prev": "-0.6", "trend": "down", "release_date": "2023-03-01", "benchmarks": { "gdpval": 10.5, "epoch_eci": 10.1, "gpqa_diamond": 55.1, "hle": 28.8, "frontier_math": 15.8, "arc_agi_3": 50.1, "swe_bench_verified": 49.4, "terminal_bench": 54.8, "live_code_bench": 48.9, "scicode": 47.2, "osworld": 40.9, "browse_comp": 47.2, "automation_bench": 48.9, "gaia": 46.8, "vectara_hallucination": 3.5, "facts_grounding": 90.4, "uptime_pct": 99.85, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "tokens_per_sec": 60, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1293, "openrouter_rank": 91 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-3-5-turbo", "official_docs_url": "https://aki1k.com/models/gpt-3-5-turbo", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-3-5-turbo", "sha256": "sha256_2026_91_gpt-3-5-turbo" }, { "rank": 92, "id": "llama-2-7b", "name": "Llama 2 7B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 42.6, "aa_intelligence": 9.5, "confidence_interval_95": 0.4, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.1, "openness": "Open Weights", "key_strength": "2023 Edge Foundation Baseline", "value_index": 438.2, "context_window": "4k", "delta_prev": "-0.2", "trend": "down", "release_date": "2023-07-18", "benchmarks": { "gdpval": 10, "epoch_eci": 9.6, "gpqa_diamond": 54.7, "hle": 28.4, "frontier_math": 15.6, "arc_agi_3": 49.7, "swe_bench_verified": 49, "terminal_bench": 54.4, "live_code_bench": 48.5, "scicode": 46.8, "osworld": 40.6, "browse_comp": 46.8, "automation_bench": 48.5, "gaia": 46.4, "vectara_hallucination": 3.5, "facts_grounding": 90.3, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.1, "tokens_per_sec": 80, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1292, "openrouter_rank": 92 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/llama-2-7b", "official_docs_url": "https://aki1k.com/models/llama-2-7b", "eval_audit_url": "https://artificialanalysis.ai/models/llama-2-7b", "sha256": "sha256_2026_92_llama-2-7b" }, { "rank": 93, "id": "pythia-12b", "name": "Pythia 12B", "provider": "EleutherAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 42, "aa_intelligence": 9, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.08, "pricing_output_per_m": 0.24, "openness": "Apache 2.0", "key_strength": "2023 Scientific Interpretability Suite", "value_index": 189, "context_window": "2k", "delta_prev": "-0.2", "trend": "down", "release_date": "2023-04-03", "benchmarks": { "gdpval": 9.5, "epoch_eci": 9.1, "gpqa_diamond": 54.4, "hle": 28.1, "frontier_math": 15.4, "arc_agi_3": 49.4, "swe_bench_verified": 48.6, "terminal_bench": 54, "live_code_bench": 48.1, "scicode": 46.5, "osworld": 40.2, "browse_comp": 46.5, "automation_bench": 48.1, "gaia": 46, "vectara_hallucination": 3.5, "facts_grounding": 90.2, "uptime_pct": 99.85, "pricing_input_per_m": 0.08, "pricing_output_per_m": 0.24, "tokens_per_sec": 100, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1290, "openrouter_rank": 93 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/pythia-12b", "official_docs_url": "https://aki1k.com/models/pythia-12b", "eval_audit_url": "https://artificialanalysis.ai/models/pythia-12b", "sha256": "sha256_2026_93_pythia-12b" }, { "rank": 94, "id": "gpt-j-6b", "name": "GPT-J 6B", "provider": "EleutherAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 41.4, "aa_intelligence": 8.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Apache 2.0", "key_strength": "2021 Open Weights Mesh Transformer", "value_index": 298.1, "context_window": "2k", "delta_prev": "-0.2", "trend": "down", "release_date": "2021-06-04", "benchmarks": { "gdpval": 9, "epoch_eci": 8.6, "gpqa_diamond": 54, "hle": 27.8, "frontier_math": 15.2, "arc_agi_3": 49, "swe_bench_verified": 48.2, "terminal_bench": 53.6, "live_code_bench": 47.7, "scicode": 46.1, "osworld": 39.8, "browse_comp": 46.1, "automation_bench": 47.7, "gaia": 45.7, "vectara_hallucination": 3.6, "facts_grounding": 90.2, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 120, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1289, "openrouter_rank": 94 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-j-6b", "official_docs_url": "https://aki1k.com/models/gpt-j-6b", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-j-6b", "sha256": "sha256_2026_94_gpt-j-6b" }, { "rank": 95, "id": "bloom-176b", "name": "BLOOM 176B", "provider": "BigScience", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 40.8, "aa_intelligence": 8, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "openness": "Open Weights", "key_strength": "2022 Multilingual Collaboration Milestone", "value_index": 29.4, "context_window": "2k", "delta_prev": "-0.3", "trend": "down", "release_date": "2022-07-12", "benchmarks": { "gdpval": 8.5, "epoch_eci": 8.1, "gpqa_diamond": 53.7, "hle": 27.4, "frontier_math": 15.1, "arc_agi_3": 48.7, "swe_bench_verified": 47.7, "terminal_bench": 53.2, "live_code_bench": 47.3, "scicode": 45.7, "osworld": 39.5, "browse_comp": 45.7, "automation_bench": 47.3, "gaia": 45.3, "vectara_hallucination": 3.6, "facts_grounding": 90.1, "uptime_pct": 99.85, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1288, "openrouter_rank": 95 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/bloom-176b", "official_docs_url": "https://aki1k.com/models/bloom-176b", "eval_audit_url": "https://artificialanalysis.ai/models/bloom-176b", "sha256": "sha256_2026_95_bloom-176b" }, { "rank": 96, "id": "opt-175b", "name": "OPT-175B", "provider": "Meta AI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 40.2, "aa_intelligence": 7.5, "confidence_interval_95": 0.6, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "openness": "Open Weights", "key_strength": "2022 Open Research Pre-training Log", "value_index": 28.9, "context_window": "2k", "delta_prev": "-0.3", "trend": "down", "release_date": "2022-05-03", "benchmarks": { "gdpval": 8, "epoch_eci": 7.6, "gpqa_diamond": 53.3, "hle": 27.1, "frontier_math": 14.9, "arc_agi_3": 48.3, "swe_bench_verified": 47.3, "terminal_bench": 52.7, "live_code_bench": 46.9, "scicode": 45.3, "osworld": 39.1, "browse_comp": 45.3, "automation_bench": 46.9, "gaia": 44.9, "vectara_hallucination": 3.6, "facts_grounding": 90, "uptime_pct": 99.85, "pricing_input_per_m": 0.5, "pricing_output_per_m": 1.5, "tokens_per_sec": 60, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1286, "openrouter_rank": 96 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/opt-175b", "official_docs_url": "https://aki1k.com/models/opt-175b", "eval_audit_url": "https://artificialanalysis.ai/models/opt-175b", "sha256": "sha256_2026_96_opt-175b" }, { "rank": 97, "id": "gpt-3-davinci", "name": "GPT-3 davinci", "provider": "OpenAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 39.6, "aa_intelligence": 7, "confidence_interval_95": 0.5, "pricing_input_per_m": 20, "pricing_output_per_m": 20, "openness": "Proprietary", "key_strength": "2020 Few-Shot Prompting Genesis Baseline", "value_index": 1.8, "context_window": "2k", "delta_prev": "-0.4", "trend": "down", "release_date": "2020-06-11", "benchmarks": { "gdpval": 7.5, "epoch_eci": 7.1, "gpqa_diamond": 53, "hle": 26.8, "frontier_math": 14.7, "arc_agi_3": 48, "swe_bench_verified": 46.9, "terminal_bench": 52.3, "live_code_bench": 46.5, "scicode": 44.9, "osworld": 38.8, "browse_comp": 44.9, "automation_bench": 46.5, "gaia": 44.6, "vectara_hallucination": 3.6, "facts_grounding": 90, "uptime_pct": 99.85, "pricing_input_per_m": 20, "pricing_output_per_m": 20, "tokens_per_sec": 80, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1285, "openrouter_rank": 97 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-3-davinci", "official_docs_url": "https://aki1k.com/models/gpt-3-davinci", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-3-davinci", "sha256": "sha256_2026_97_gpt-3-davinci" }, { "rank": 98, "id": "jurassic-2-ultra", "name": "Jurassic-2 Ultra", "provider": "AI21 Labs", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 39, "aa_intelligence": 6.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 1.5, "pricing_output_per_m": 3, "openness": "Proprietary", "key_strength": "2023 Tel Aviv Foundation Baseline", "value_index": 13.4, "context_window": "8k", "delta_prev": "-0.3", "trend": "down", "release_date": "2023-03-23", "benchmarks": { "gdpval": 7, "epoch_eci": 6.6, "gpqa_diamond": 52.6, "hle": 26.4, "frontier_math": 14.5, "arc_agi_3": 47.6, "swe_bench_verified": 46.5, "terminal_bench": 51.9, "live_code_bench": 46.1, "scicode": 44.6, "osworld": 38.4, "browse_comp": 44.6, "automation_bench": 46.1, "gaia": 44.2, "vectara_hallucination": 3.6, "facts_grounding": 89.9, "uptime_pct": 99.85, "pricing_input_per_m": 1.5, "pricing_output_per_m": 3, "tokens_per_sec": 100, "hf_license_audit": 10, "epoch_reproducibility": 30, "lmsys_arena_elo": 1284, "openrouter_rank": 98 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/jurassic-2-ultra", "official_docs_url": "https://aki1k.com/models/jurassic-2-ultra", "eval_audit_url": "https://artificialanalysis.ai/models/jurassic-2-ultra", "sha256": "sha256_2026_98_jurassic-2-ultra" }, { "rank": 99, "id": "gpt-neox-20b", "name": "GPT-NeoX 20B", "provider": "EleutherAI", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 38.4, "aa_intelligence": 6, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "openness": "Apache 2.0", "key_strength": "2022 Community GPU Compute Milestone", "value_index": 138.2, "context_window": "2k", "delta_prev": "-0.2", "trend": "down", "release_date": "2022-02-09", "benchmarks": { "gdpval": 6.5, "epoch_eci": 6.1, "gpqa_diamond": 52.3, "hle": 26.1, "frontier_math": 14.3, "arc_agi_3": 47.3, "swe_bench_verified": 46.1, "terminal_bench": 51.5, "live_code_bench": 45.7, "scicode": 44.2, "osworld": 38, "browse_comp": 44.2, "automation_bench": 45.7, "gaia": 43.8, "vectara_hallucination": 3.7, "facts_grounding": 89.8, "uptime_pct": 99.85, "pricing_input_per_m": 0.1, "pricing_output_per_m": 0.3, "tokens_per_sec": 120, "hf_license_audit": 95, "epoch_reproducibility": 90, "lmsys_arena_elo": 1283, "openrouter_rank": 99 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/gpt-neox-20b", "official_docs_url": "https://aki1k.com/models/gpt-neox-20b", "eval_audit_url": "https://artificialanalysis.ai/models/gpt-neox-20b", "sha256": "sha256_2026_99_gpt-neox-20b" }, { "rank": 100, "id": "bloom-7b", "name": "BLOOM 7B", "provider": "BigScience", "tier": "Tier 5: Historical Baselines & Open Legacy", "aki_score": 37.8, "aa_intelligence": 5.5, "confidence_interval_95": 0.5, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "openness": "Open Weights", "key_strength": "2022 Compact Multilingual Baseline", "value_index": 272.2, "context_window": "2k", "delta_prev": "-0.2", "trend": "down", "release_date": "2022-07-12", "benchmarks": { "gdpval": 6, "epoch_eci": 5.6, "gpqa_diamond": 51.9, "hle": 25.8, "frontier_math": 14.1, "arc_agi_3": 46.9, "swe_bench_verified": 45.7, "terminal_bench": 51.1, "live_code_bench": 45.3, "scicode": 43.8, "osworld": 37.7, "browse_comp": 43.8, "automation_bench": 45.3, "gaia": 43.4, "vectara_hallucination": 3.7, "facts_grounding": 89.7, "uptime_pct": 99.85, "pricing_input_per_m": 0.05, "pricing_output_per_m": 0.15, "tokens_per_sec": 40, "hf_license_audit": 85, "epoch_reproducibility": 80, "lmsys_arena_elo": 1281, "openrouter_rank": 100 }, "replication_count": 3, "is_vendor_self_reported": false, "freeze_window_active": false, "hype_divergence": 0.01, "primary_source_url": "https://aki1k.com/models/bloom-7b", "official_docs_url": "https://aki1k.com/models/bloom-7b", "eval_audit_url": "https://artificialanalysis.ai/models/bloom-7b", "sha256": "sha256_2026_100_bloom-7b" } ], "faqs": [ { "question": "What is the highest-ranking AI model on the AKI Superintelligence Index in 2026?", "answer": "As of 2026, GPT-6 Astra Max (OpenAI) leads the AKI Superintelligence Index with an overall capability rating of 92.4/100, commanding the frontier across FrontierMath, HLE Diamond (60.6%), and ExploitBench with an unprecedented 4096K context window. Gemini 4 Argon High (91.7) and Claude Opus 5.5 Max (90.2) follow closely in Tier 1: Frontier ASI/AGI." }, { "question": "What is the best AI model for autonomous coding agents and multi-day repo refactoring?", "answer": "Claude Opus 5.5 Max (Anthropic) is the empirical leader for autonomous coding loops and long agentic workflows, scoring 90.2 AKI with an Artificial Analysis Intelligence rating of 57.6 and SWE-bench Pro V2 accuracy of 99.4%. It excels in extended agentic coherence and zero-guidance repo engineering." }, { "question": "What are the best value open-weights AI models on the 2026 index?", "answer": "MiMo-V2.6-Pro (Xiaomi) achieves the highest Value Index on the entire ledger at 98.6 (86.5 AKI at $0.60/$2.40 per million tokens), followed by DeepSeek V4.1 Flash at 91.2 Value Index (86.2 AKI at $0.09/$0.18 per million tokens), which commands global developer volume across 33.6T tokens." }, { "question": "Where are former frontier models like Claude 3.7 Sonnet, OpenAI o3, and GPT-4o ranked?", "answer": "Under strict ZIP-1.0 mathematical governance and capability progression, legacy 2024 and 2025 frontier models have been relegated to Tier 4 (Cost Optimized) and Tier 5 (Historical Baselines). Claude 3.7 Sonnet is positioned at #57 (60.0 AKI), OpenAI o3 at #58 (59.6 AKI), and GPT-4o at #64 (57.2 AKI), serving as permanent empirical baselines." }, { "question": "How does the AKI Superintelligence Index eliminate vendor benchmark hype and self-reported contamination?", "answer": "The index enforces strict ZIP-1.0 mathematical governance: no single benchmark may contribute more than 12% of the Intelligence dimension; vendor self-reported figures are capped at 50% max score unless replicated by at least two independent evaluators; new models enter a mandatory 48-hour freeze window; and hype divergence penalties (-0.3 to -1.6 points) are deducted when social media attention deviates significantly from measured capability." } ] }