Preserved baseline cohort of landmark earlier AI models (DeepSeek-R1, original Claude 3.5 Sonnet, OpenAI o3-mini, Llama 3.3 70B, SmolLM2, Whisper v3 turbo) with verified original timestamps and mathematical recency decay.
| Rank | Release Date | Model & Lab | Category | Architecture | Score | Action |
|---|---|---|---|---|---|---|
| #101 | Jan 20, 2025 |
DeepSeek-R1 (671B MoE & Distillations)
DeepSeek AI
|
Frontier Models & Flagships | Mixture-of-Experts (MoE) with Multi-Head Latent Attention (MLA) | 82.3 | Try β |
| #102 | Oct 22, 2024 |
Claude 3.5 Sonnet (Original October Release)
Anthropic
|
Consumer Agents & Platforms | Autoregressive Transformer with Computer Use Tool API | 71.8 | Try β |
| #103 | Jan 31, 2025 |
OpenAI o3-mini Reasoning Engine
OpenAI
|
Frontier Models & Flagships | Reinforcement Learning CoT Transformer | 79.7 | Try β |
| #104 | Dec 6, 2024 |
Llama 3.3 70B Instruct (December 2024 Launch)
Meta AI
|
Open-Weights & MoE Engines | Dense Auto-regressive Transformer (GQA 8-KV heads) | 68.8 | Try β |
| #105 | Nov 4, 2024 |
SmolLM2 1.7B (November 2024 Initial Release)
Hugging Face
|
Open-Weights & MoE Engines | Compact Dense Transformer for Edge Devices | 62.8 | Try β |
| #106 | Oct 1, 2024 |
Whisper large-v3-turbo (October 2024 Initial Release)
OpenAI
|
Multimodal, Video & Speech | Encoder-Decoder Speech Transformer | 63.1 | Try β |
| #107 | Dec 10, 2024 |
Ollama 0.5.x Local Model Engine Baseline
Ollama Project
|
Coding Agents & Dev Tools | llama.cpp Core with Custom Go HTTP Server | 64.3 | Try β |
| #108 | Nov 12, 2024 |
Qwen 2.5-Coder-32B Instruct Baseline
Alibaba Cloud
|
Coding Agents & Dev Tools | Dense Decoder-Only Transformer (GQA) | 70.8 | Try β |
| #109 | May 29, 2024 |
Codestral 22B Code Specialist
Mistral AI
|
Coding Agents & Dev Tools | Dense Auto-regressive Transformer with FIM | 59.1 | Try β |
| #110 | Aug 1, 2024 |
FLUX.1 [dev] Open-Weights Imaging
Black Forest Labs
|
Multimodal, Video & Speech | 12B Flow Matching Transformer (DiT) | 68.7 | Try β |
| #111 | Jun 27, 2024 |
Gemma 2 27B Open Foundation
Google DeepMind
|
Open-Weights & MoE Engines | Dense Transformer with Sliding Window & GQA | 63.3 | Try β |
| #112 | Jun 17, 2024 |
Runway Gen-3 Alpha Cinematic Video
Runway
|
Multimodal, Video & Speech | Diffusion-Transformer (DiT) Video Architecture | 63.2 | Try β |
| #113 | May 29, 2024 |
Suno v3.5 Generative Music Engine
Suno AI
|
Multimodal, Video & Speech | Discrete Audio Diffusion Transformer with Vocoder | 61.0 | Try β |
| #114 | Jul 31, 2024 |
Udio 1.5 High-Definition Music
Udio
|
Multimodal, Video & Speech | High-Fidelity Audio Latent Diffusion Model | 63.5 | Try β |
| #115 | Feb 15, 2024 |
Gemini 1.5 Pro (Original 1M Window Release)
Google DeepMind
|
Frontier Models & Flagships | Sparse Mixture-of-Experts with 1M-2M Context Attention | 62.5 | Try β |