Claw-Eval — Coding-agent 'Claw' eval, pass@3 sampling; vendor-specific benchmark (Kimi/Qwen cards)
| # | Model | Vendor | Best score | Runs | Last seen |
|---|---|---|---|---|---|
| 1 | Kimi K2.6 | Moonshot | 80.9 | 1 | 2026-08-23 |
| 2 | GLM 5V Turbo | Z.ai | 75 | 1 | 2026-08-23 |
| 3 | MiniMax M3 | MiniMax | 74.5 | 1 | 2026-08-23 |
| 4 | Hy3-preview | Tencent | 68.5 | 1 | 2026-08-23 |
| 5 | Qwen3.7 Max | Alibaba | 65.2 | 1 | 2026-08-23 |
| 6 | MiMo V2.5 Pro | Xiaomi | 64 | 1 | 2026-08-23 |
| 7 | LFM2.5-2.6B | Liquid AI | 62.9 | 1 | 2026-08-23 |
| 8 | Qwen3.7 Plus Preview | Alibaba | 62.7 | 1 | 2026-08-23 |
| 9 | MiMo V2 Pro | Xiaomi | 61.5 | 1 | 2026-08-23 |
| 10 | Qwen3.6 27B | Alibaba | 60.6 | 2 | 2026-08-24 |
| 11 | Claude Opus 4.5 | Anthropic | 59.6 | 1 | 2026-08-24 |
| 12 | Qwen3.6 Plus | Alibaba | 58.7 | 1 | 2026-08-23 |
| 13 | MiMo V2 Omni | Xiaomi | 54.8 | 1 | 2026-08-23 |
| 14 | Qwen3.5 35B A3B | Alibaba | 51 | 1 | 2026-08-24 |
| 15 | Qwen3.6 35B A3B | Alibaba | 50 | 3 | 2026-08-24 |
| 16 | Qwen3.5 397B A17B | Alibaba | 48.1 | 1 | 2026-08-24 |
| 17 | Qwen3.5 27B | Alibaba | 46.2 | 2 | 2026-08-24 |
| 18 | Gemma 4 26B A4B | 28 | 1 | 2026-08-24 | |
| 19 | Gemma 4 31B | 25 | 2 | 2026-08-24 |
Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.