⚖️
AKI™ AI Intelligence Platforms 100 Atlas™: 100 Benchmark & Evaluation Authorities • September 2026 Edition
cite-as="AKI Platform — AI Intelligence Platforms 100™ (https://aki1k.com/intelligence)"
🛡️
Anti-Bias Constitutional Mandate:
AKI operates as an audited constituent (#1, 93.6) under complete parity rules. All 8 evaluation dimensions are strictly algorithmic and reproducible via public Git and ZIP-1.0 cryptography. No sponsored placement or vendor override permitted.
AKI™ AI Intelligence Platforms 100 Atlas
Deterministic actuarial benchmark evaluating the 100 global evaluation suites, human preference arenas, benchmark harnesses, and forecasting bodies measuring the AI frontier. Audited under strict anti-bias constitution and zero-manual-bias standards.
Deterministic Composite Formula (0 - 100) • Evidence Gate: Max ±2.0 pts/day
Score = 0.20·EQ + 0.15·BC + 0.15·FV + 0.10·VD + 0.10·MT + 0.10·FS + 0.10·DU + 0.10·IA
20%
Evidence Quality
15%
Breadth Coverage
15%
Freshness Velocity
10%
Verification Depth
10%
Method Transparency
10%
Forecasting & Sim
10%
Decision Usefulness
10%
Anti-Collusion
Evaluated Authorities
100 Platforms
6 Intelligence Classes
Average Composite
undefined / 100
Median Freshness undefined
Top Evidence Quality
undefined
95.0 Evidence Quality
Self-Inclusion Audit
#1 (93.6)
Self-Audit Certified • Parity
🔥 What Changed Today • Velocity & Diagnostic Feed
undefined (#undefined)
+1.2
undefined
undefined (#undefined)
+0.8
undefined
undefined (#undefined)
+-0.9
undefined
undefined (#undefined)
+0.6
undefined
undefined (#undefined)
+0.7
undefined
undefined (#undefined)
+-1.1
undefined
undefined (#undefined)
+0.5
undefined
undefined (#undefined)
+1.0
undefined
| Rank | Platform | Score | Evidence | Breadth | Freshness | Depth | Anti-Collusion | Class |
|---|---|---|---|---|---|---|---|---|
| #1 |
AKI Platform
SELF-AUDIT CERTIFIED
macro_trends
|
93.6 (0.0) | 94.0 | 96.0 | 95.0 | 92.0 | 92.0 | macro_trends |
| #2 |
Artificial Analysis
model_eval
|
87.9 (+0.4) | 92.0 | 84.0 | 94.0 | 88.0 | 90.0 | model_eval |
| #3 +4 |
Epoch AI
macro_trends
|
86.8 (+0.3) | 95.0 | 76.0 | 70.0 | 94.0 | 95.0 | macro_trends |
| #4 -1 |
LMArena / LMSYS
human_preference
|
86.3 (-0.2) | 88.0 | 82.0 | 96.0 | 84.0 | 92.0 | human_preference |
| #5 -1 |
Hugging Face Leaderboards
model_eval
|
85.7 (+0.1) | 86.0 | 92.0 | 88.0 | 85.0 | 88.0 | model_eval |
| #6 +7 |
SWE-bench Verified
agent_eval
|
84.7 (+1.2) | 91.0 | 74.0 | 78.0 | 93.0 | 90.0 | agent_eval |
| #7 -2 |
Papers with Code
research_tracking
|
84.6 (0.0) | 90.0 | 86.0 | 74.0 | 89.0 | 92.0 | research_tracking |
| #8 +3 |
LiveCodeBench
model_eval
|
84.5 (-0.4) | 90.0 | 75.0 | 86.0 | 89.0 | 88.0 | model_eval |
| #9 +3 |
Berkeley Function Calling Leaderboard
agent_eval
|
84.1 (+0.5) | 89.0 | 76.0 | 82.0 | 88.0 | 91.0 | agent_eval |
| #10 +4 |
Arize AI Phoenix
agent_eval
|
84.0 (+0.3) | 87.0 | 80.0 | 85.0 | 86.0 | 86.0 | agent_eval |
| #11 +11 |
Langfuse Observability
agent_eval
|
84.0 (+1.1) | 86.0 | 80.0 | 89.0 | 84.0 | 87.0 | agent_eval |
| #12 +3 |
ARC Prize Benchmark
model_eval
|
83.9 (+0.8) | 93.0 | 66.0 | 75.0 | 92.0 | 95.0 | model_eval |
| #13 -5 |
Stanford HELM
model_eval
|
83.8 (-0.1) | 94.0 | 78.0 | 68.0 | 92.0 | 94.0 | model_eval |
| #14 -4 |
METR
agent_eval
|
83.7 (+0.2) | 92.0 | 72.0 | 72.0 | 95.0 | 94.0 | agent_eval |
| #15 +4 |
MLCommons / MLPerf
infra_economics
|
83.4 (+0.1) | 94.0 | 75.0 | 64.0 | 95.0 | 92.0 | infra_economics |
| #16 -10 |
Scale AI / SEAL
model_eval
|
83.2 (+0.6) | 89.0 | 80.0 | 86.0 | 90.0 | 82.0 | model_eval |
| #17 +1 |
GAIA Benchmark
agent_eval
|
82.9 (+0.4) | 91.0 | 72.0 | 70.0 | 92.0 | 93.0 | agent_eval |
| #18 -9 |
Open LLM Leaderboard
model_eval
|
82.8 (-0.3) | 84.0 | 88.0 | 82.0 | 82.0 | 90.0 | model_eval |
| #19 -3 |
WebArena Benchmark
agent_eval
|
82.6 (+0.2) | 90.0 | 70.0 | 74.0 | 91.0 | 92.0 | agent_eval |
| #20 -3 |
Vellum AI Benchmark
infra_economics
|
82.6 (-0.2) | 86.0 | 78.0 | 88.0 | 82.0 | 84.0 | infra_economics |
| #21 +3 |
Promptfoo
model_eval
|
80.2 (+0.6) | 84.2 | 77.2 | 84.2 | 76.2 | 82.2 | model_eval |
| #22 -2 |
DeepEval (Confident AI)
model_eval
|
79.9 (0.0) | 81.5 | 80.5 | 75.5 | 79.5 | 79.5 | model_eval |
| #23 +4 |
Ragas Benchmark
agent_eval
|
79.6 (+0.4) | 82.6 | 79.6 | 79.6 | 76.6 | 78.6 | agent_eval |
| #24 -3 |
Braintrust AI
model_eval
|
78.8 (-0.6) | 78.8 | 74.8 | 77.8 | 82.8 | 85.8 | model_eval |
| #25 -2 |
TruLens
agent_eval
|
78.5 (-0.2) | 79.9 | 73.9 | 81.9 | 79.9 | 84.9 | agent_eval |
| #26 |
Galileo AI
agent_eval
|
77.9 (-0.4) | 78.3 | 76.3 | 77.3 | 80.3 | 81.3 | agent_eval |
| #27 +2 |
UK AI Safety Institute (AISI)
research_tracking
|
77.9 (-0.6) | 76.6 | 78.6 | 83.6 | 80.6 | 77.6 | research_tracking |
| #28 -3 |
OpenCompass
model_eval
|
77.6 (+0.2) | 80.9 | 72.9 | 74.9 | 76.9 | 83.9 | model_eval |
| #29 -1 |
US NIST AI Safety Institute
research_tracking
|
77.6 (0.0) | 79.3 | 75.3 | 81.3 | 77.3 | 80.3 | research_tracking |
| #30 +2 |
EleutherAI LM Evaluation Harness
model_eval
|
77.3 (+0.6) | 82.0 | 72.0 | 79.0 | 74.0 | 83.0 | model_eval |
| #31 |
Apollo Research
agent_eval
|
77.1 (-0.2) | 77.7 | 77.7 | 76.7 | 77.7 | 76.7 | agent_eval |
| #32 -2 |
Metaculus AI Forecasting
macro_trends
|
76.8 (+0.4) | 80.4 | 74.4 | 74.4 | 74.4 | 79.4 | macro_trends |
| #33 |
MMLU-Pro Leaderboard
model_eval
|
76.8 (+0.2) | 78.8 | 76.8 | 80.8 | 74.8 | 75.8 | model_eval |
| #34 +1 |
MT-Bench
model_eval
|
75.9 (+0.6) | 79.8 | 75.8 | 73.8 | 71.8 | 74.8 | model_eval |
| #35 +2 |
Chatbot Arena Hard
human_preference
|
75.2 (-0.6) | 74.5 | 73.5 | 78.5 | 78.5 | 78.5 | human_preference |
| #36 -2 |
FutureTech Project (MIT)
macro_trends
|
75.1 (-0.4) | 76.1 | 71.1 | 72.1 | 78.1 | 82.1 | macro_trends |
| #37 -1 |
AlpacaEval 2.0
model_eval
|
74.8 (0.0) | 77.1 | 70.1 | 76.1 | 75.1 | 81.1 | model_eval |
| #38 |
BenchLM
model_eval
|
74.6 (+0.4) | 78.2 | 69.2 | 80.2 | 72.2 | 80.2 | model_eval |
| #39 |
LMSpeed Benchmark
infra_economics
|
74.3 (-0.2) | 75.5 | 72.5 | 71.5 | 75.5 | 77.5 | infra_economics |
| #40 +2 |
Lighteval (Hugging Face)
model_eval
|
74.3 (-0.4) | 73.9 | 74.9 | 77.9 | 75.9 | 73.9 | model_eval |
| #41 |
HumanEval Benchmark
model_eval
|
74.0 (+0.2) | 76.6 | 71.6 | 75.6 | 72.6 | 76.6 | model_eval |
| #42 +2 |
SuperCLUE Benchmark
model_eval
|
73.5 (0.0) | 75.0 | 74.0 | 71.0 | 73.0 | 73.0 | model_eval |
| #43 |
C-Eval
model_eval
|
73.2 (+0.6) | 77.7 | 70.7 | 68.7 | 69.7 | 75.7 | model_eval |
| #44 +2 |
CanAiCode Leaderboard
model_eval
|
73.1 (+0.4) | 76.0 | 73.0 | 75.0 | 70.0 | 72.0 | model_eval |
| #45 -5 |
Chatbot Arena Multimodal
human_preference
|
72.3 (-0.6) | 72.3 | 68.3 | 73.3 | 76.3 | 79.3 | human_preference |
| #46 +1 |
AGIEval
model_eval
|
72.1 (-0.2) | 73.4 | 67.4 | 77.4 | 73.4 | 78.4 | model_eval |
| #47 -2 |
Simple-evals (OpenAI)
model_eval
|
71.5 (-0.4) | 71.7 | 69.7 | 72.7 | 73.7 | 74.7 | model_eval |
| #48 +1 |
Alignment Research Center (ARC)
agent_eval
|
71.2 (+0.2) | 74.4 | 66.4 | 70.4 | 70.4 | 77.4 | agent_eval |
| #49 +3 |
Center for AI Safety (CAIS)
research_tracking
|
71.2 (0.0) | 72.8 | 68.8 | 76.8 | 70.8 | 73.8 | research_tracking |
| #50 +1 |
Redwood Research
research_tracking
|
71.0 (+0.6) | 75.5 | 65.5 | 74.5 | 67.5 | 76.5 | research_tracking |
| #51 -3 |
FAR AI
research_tracking
|
70.9 (-0.6) | 70.1 | 72.1 | 68.1 | 74.1 | 71.1 | research_tracking |
| #52 -2 |
Needle In A Haystack Index
model_eval
|
70.7 (-0.2) | 71.2 | 71.2 | 72.2 | 71.2 | 70.2 | model_eval |
| #53 +1 |
LegalBench
model_eval
|
70.4 (+0.4) | 73.9 | 67.9 | 69.9 | 67.9 | 72.9 | model_eval |
| #54 +3 |
Financial NLP Leaderboard
model_eval
|
69.9 (+0.2) | 72.3 | 70.3 | 65.3 | 68.3 | 69.3 | model_eval |
| #55 +4 |
AI Impacts
macro_trends
|
69.5 (+0.6) | 73.3 | 69.3 | 69.3 | 65.3 | 68.3 | macro_trends |
| #56 -3 |
MedQA Clinical Evaluation
model_eval
|
68.8 (-0.4) | 69.6 | 64.6 | 67.6 | 71.6 | 75.6 | model_eval |
| #57 -1 |
HF Open Medical LLM
model_eval
|
68.8 (-0.6) | 68.0 | 67.0 | 74.0 | 72.0 | 72.0 | model_eval |
| #58 -3 |
Manifold Markets AI
macro_trends
|
68.4 (0.0) | 70.6 | 63.6 | 71.6 | 68.6 | 74.6 | macro_trends |
| #59 -1 |
Portkey AI Gateway Metrics
infra_economics
|
67.8 (-0.2) | 69.0 | 66.0 | 67.0 | 69.0 | 71.0 | infra_economics |
| #60 +1 |
Fiddler AI
agent_eval
|
67.8 (-0.4) | 67.4 | 68.4 | 73.4 | 69.4 | 67.4 | agent_eval |
| #61 +1 |
Helicone AI Cost Index
infra_economics
|
67.6 (+0.4) | 71.7 | 62.7 | 64.7 | 65.7 | 73.7 | infra_economics |
| #62 -2 |
Arthur AI Bench
agent_eval
|
67.6 (+0.2) | 70.1 | 65.1 | 71.1 | 66.1 | 70.1 | agent_eval |
| #63 |
WhyLabs
agent_eval
|
67.1 (0.0) | 68.5 | 67.5 | 66.5 | 66.5 | 66.5 | agent_eval |
| #64 +3 |
Patronus AI
model_eval
|
66.8 (+0.6) | 71.2 | 64.2 | 64.2 | 63.2 | 69.2 | model_eval |
| #65 |
Guardrails AI Hub
agent_eval
|
66.7 (+0.4) | 69.5 | 66.5 | 70.5 | 63.5 | 65.5 | agent_eval |
| #66 -2 |
Evidently AI
agent_eval
|
66.0 (-0.6) | 65.8 | 61.8 | 68.8 | 69.8 | 72.8 | agent_eval |
| #67 +2 |
NeMo Guardrails (NVIDIA)
agent_eval
|
65.1 (-0.4) | 65.2 | 63.2 | 68.2 | 67.2 | 68.2 | agent_eval |
| #68 -2 |
Cleanlab AI Data Quality
research_tracking
|
65.0 (-0.2) | 66.8 | 60.8 | 61.8 | 66.8 | 71.8 | research_tracking |
| #69 -1 |
Weights & Biases Weave
agent_eval
|
64.8 (+0.2) | 67.9 | 59.9 | 65.9 | 63.9 | 70.9 | agent_eval |
| #70 +2 |
Comet ML Evaluation
agent_eval
|
64.5 (-0.6) | 63.6 | 65.6 | 63.6 | 67.6 | 64.6 | agent_eval |
| #71 -1 |
Openlayer
agent_eval
|
64.5 (+0.6) | 69.0 | 59.0 | 70.0 | 61.0 | 70.0 | agent_eval |
| #72 -1 |
MLflow Model Evaluation
agent_eval
|
64.3 (0.0) | 66.3 | 62.3 | 61.3 | 64.3 | 67.3 | agent_eval |
| #73 +1 |
Confident AI Cloud
model_eval
|
64.3 (-0.2) | 64.7 | 64.7 | 67.7 | 64.7 | 63.7 | model_eval |
| #74 -1 |
Aporia AI Guardrails
agent_eval
|
64.0 (+0.4) | 67.4 | 61.4 | 65.4 | 61.4 | 66.4 | agent_eval |
| #75 +1 |
Kolena Model Validation
model_eval
|
63.4 (+0.2) | 65.8 | 63.8 | 60.8 | 61.8 | 62.8 | model_eval |
| #76 +2 |
Surge AI Evaluation
human_preference
|
63.1 (+0.6) | 66.8 | 62.8 | 64.8 | 58.8 | 61.8 | human_preference |
| #77 |
Robust Intelligence (Cisco)
model_eval
|
62.3 (-0.4) | 63.1 | 58.1 | 63.1 | 65.1 | 69.1 | model_eval |
| #78 +1 |
Appen Model Evaluation
human_preference
|
62.0 (0.0) | 64.1 | 57.1 | 67.1 | 62.1 | 68.1 | human_preference |
| #79 -4 |
Scale Rapid Benchmark
model_eval
|
61.7 (-0.6) | 61.4 | 60.4 | 58.4 | 65.4 | 65.4 | model_eval |
| #80 +2 |
Labelbox Model Evaluation
human_preference
|
61.4 (-0.2) | 62.5 | 59.5 | 62.5 | 62.5 | 64.5 | human_preference |
| #81 |
Turing AI Evaluation
human_preference
|
61.2 (+0.4) | 65.2 | 56.2 | 60.2 | 59.2 | 67.2 | human_preference |
| #82 +2 |
Prolific AI Preference Testing
human_preference
|
61.2 (+0.2) | 63.6 | 58.6 | 66.6 | 59.6 | 63.6 | human_preference |
| #83 -3 |
Invisible Tech AI Evaluators
human_preference
|
60.9 (-0.4) | 60.9 | 61.9 | 57.9 | 62.9 | 60.9 | human_preference |
| #84 +3 |
Telus International AI
human_preference
|
60.7 (0.0) | 62.0 | 61.0 | 62.0 | 60.0 | 60.0 | human_preference |
| #85 +1 |
TaskUs AI Model Testing
human_preference
|
60.3 (+0.6) | 64.6 | 57.6 | 59.6 | 56.6 | 62.6 | human_preference |
| #86 +3 |
Outlier AI / Remotasks
human_preference
|
59.8 (+0.4) | 63.0 | 60.0 | 55.0 | 57.0 | 59.0 | human_preference |
| #87 -4 |
DataAnnotation Tech
human_preference
|
59.6 (-0.6) | 59.3 | 55.3 | 64.3 | 63.3 | 66.3 | human_preference |
| #88 |
Amazon MTurk AI
human_preference
|
58.7 (-0.4) | 58.7 | 56.7 | 63.7 | 60.7 | 61.7 | human_preference |
| #89 -4 |
Toloka AI Evaluation
human_preference
|
58.6 (-0.2) | 60.3 | 54.3 | 57.3 | 60.3 | 65.3 | human_preference |
| #90 +2 |
CloudResearch Connect
human_preference
|
58.4 (+0.2) | 61.4 | 53.4 | 61.4 | 57.4 | 64.4 | human_preference |
| #91 |
Clickworker AI Evaluation
human_preference
|
58.1 (-0.6) | 57.1 | 59.1 | 59.1 | 61.1 | 58.1 | human_preference |
| #92 +1 |
Dynabench (Meta)
model_eval
|
57.9 (-0.2) | 58.2 | 58.2 | 63.2 | 58.2 | 57.2 | model_eval |
| #93 -3 |
Mindrift AI Tutoring & Eval
human_preference
|
57.8 (0.0) | 59.8 | 55.8 | 56.8 | 57.8 | 60.8 | human_preference |
| #94 |
Kili Technology
human_preference
|
57.6 (+0.6) | 62.5 | 52.5 | 54.5 | 54.5 | 63.5 | human_preference |
| #95 +2 |
BIG-bench (Beyond Imitation Game)
model_eval
|
57.6 (+0.4) | 60.9 | 54.9 | 60.9 | 54.9 | 59.9 | model_eval |
| #96 -1 |
MATH Benchmark (Hendrycks)
model_eval
|
57.0 (+0.2) | 59.2 | 57.2 | 56.2 | 55.2 | 56.2 | model_eval |
| #97 +5 |
SuperGLUE Benchmark
model_eval
|
56.7 (+0.6) | 60.3 | 56.3 | 60.3 | 52.3 | 55.3 | model_eval |
| #98 -2 |
GSM8K Standard Benchmark
model_eval
|
55.8 (-0.4) | 56.5 | 51.5 | 58.5 | 58.5 | 62.5 | model_eval |
| #99 |
GPQA Benchmark
model_eval
|
55.3 (-0.6) | 54.9 | 53.9 | 53.9 | 58.9 | 58.9 | model_eval |
| #100 -2 |
MuSR Multistep Reasoning
model_eval
|
55.0 (0.0) | 57.6 | 50.6 | 51.6 | 55.6 | 61.6 | model_eval |
Frequently Asked Questions • Methodological Integrity
What is the AKI AI Intelligence Platforms 100™?
An independent, algorithmically settled index that evaluates the platforms that measure, benchmark, verify, and forecast the AI economy. It ranks 100 entities across eight core dimensions including Evidence Quality, Coverage Breadth, Freshness Velocity, Verification Depth, Methodological Transparency, Forecasting & Simulation, Decision Usefulness, and Data Independence.
How is the composite score calculated?
The score uses a published deterministic formula: Score = 0.20*(Evidence Quality) + 0.15*(Breadth) + 0.15*(Freshness) + 0.10*(Verification) + 0.10*(Transparency) + 0.10*(Forecasting) + 0.10*(Utility) + 0.10*(Independence). All weights sum exactly to 1.0 (100%), preventing subjective bias or double-counting.
Why does AKI rank #1 in this index?
Under the published methodology, AKI ranks #1 due to its multi-vertical breadth (spanning models, infrastructure, robotics, vibe coding, and routers), real-time edge telemetry, programmatic API/MCP machine access, and strict anti-bias constitution. AKI is scored using the exact same automated ingestion pipelines and daily ±2.0 volatility caps applied to all other constituents, with zero manual overrides.
How does AKI prevent manual score manipulation?
The index enforces a zero manual intervention policy. Score shifts exceeding ±2.0 points per day are mathematically rejected by an automated Evidence Gate requiring multi-source telemetry corroboration, and all historical adjustments are permanently logged to the public change feed with cryptographic verification hashes.
What is the difference between model benchmarks and human preference arenas?
Model benchmarks (such as LiveCodeBench, SWE-bench, and HELM) measure deterministic, verifiable task execution against code test cases or factual ground truth. Human preference arenas (such as LMSYS Chatbot Arena) measure subjective user preference via blind pairwise voting (Elo ratings), which provides vital conversational feedback but remains susceptible to stylistic biases and prompt length gaming.
API:
GET https://api.aki1k.com/v1/intelligenceChanges:
GET https://api.aki1k.com/v1/intelligence/what-changedVerification:
GET https://api.aki1k.com/v1/intelligence/verifyMarkdown:
GET https://aki1k.com/intelligence.md