| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Claude Fable 5 | Anthropic | 93.3 | 1 | 2026-07-16 |
| 2 | Gemini 3.1 Pro | 92.7 | 1 | 2026-07-16 | |
| 3 | GPT-5.6 Sol | OpenAI | 91.8 | 1 | 2026-07-16 |
| 4 | Qwen3.5 397B A17B | Alibaba | 90 | 1 | 2026-08-24 |
| 5 | DeepSeek-V4-Pro | DeepSeek | 89.3 | 1 | 2026-07-16 |
| 6 | Gemini 2.5 Pro Preview 06.05 (32k think) | 89.2 | 1 | 2026-08-24 | |
| 7 | GLM-5.2 Full Open Source | Z.ai | 89.2 | 1 | 2026-07-16 |
| 8 | Inkling-Small | Thinking Machines | 88.7 | 3 | 2026-08-24 |
| 9 | Inkling | Thinking Machines | 88.7 | 2 | 2026-07-16 |
| 10 | Gemini 2.5 Pro | 88.6 | 1 | 2026-08-23 | |
| 11 | DeepSeek-V4-Flash | DeepSeek | 88.4 | 1 | 2026-08-24 |
| 12 | Kimi K2.6 | Moonshot | 88.4 | 1 | 2026-07-16 |
| 13 | Gemini 2.5 Flash | 88.4 | 1 | 2026-08-23 | |
| 14 | Kimi K2.5 | Moonshot | 84 | 1 | 2026-07-16 |
| 15 | MiniMax M2.7 | MiniMax | 83.9 | 1 | 2026-08-24 |
| 16 | MiMo V2.5 Pro | Xiaomi | 83.5 | 1 | 2026-08-24 |
| 17 | Gemini 2.5 Flash Lite | 81.1 | 1 | 2026-08-23 | |
| 18 | Gemma 3 27B | 75.1 | 1 | 2026-08-23 | |
| 19 | Gemma 3 12B | 69.5 | 1 | 2026-08-23 | |
| 20 | Qwen3.5 2B | Alibaba | 64.5 | 1 | 2026-08-15 |
| 21 | Gemma 3 4B | 54.5 | 1 | 2026-08-23 | |
| 22 | Qwen3.5 0.8B | Alibaba | 53.3 | 1 | 2026-08-15 |
| 23 | Gemma 3 1B | 34.2 | 1 | 2026-08-23 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.