Official benchmarks map / WildClawBench
Qwen/Qwen3.8-2.4T-A95B is ranked first on the WildClawBench leaderboard with 708, 5.0% ahead of moonshotai/Kimi-K2.7-Code. Read live from the Hugging Face API on November 6, 2026.
| # | Model | Organization | Score |
|---|---|---|---|
| 1 | Qwen/Qwen3.8-2.4T-A95B | Qwen | 708 |
| 2 | moonshotai/Kimi-K2.7-Code | moonshotai | 674 |
| 3 | deepseek-ai/DeepSeek-V4-Pro | deepseek-ai | 605 |
| 4 | MiniMaxAI/MiniMax-M2.7 | MiniMaxAI | 551 |
| 5 | MiniMaxAI/MiniMax-M2.5 | MiniMaxAI | 542 |
| 6 | internlm/Intern-S2-Preview-397B | internlm | 541 |
| 7 | Qwen/Qwen3.8-27B | Qwen | 516 |
| 8 | zai-org/GLM-5.1 | zai-org | 515 |
| 9 | moonshotai/Kimi-K3 | moonshotai | 488 |
| 10 | Qwen/Qwen3.5-397B-A17B | Qwen | 459 |
WildClawBench is a dataset that Hugging Face marks as an official benchmark. Its leaderboard is built automatically from the .eval_results files that model repositories publish, so entries are mostly self-reported by model authors and evaluation settings differ between submissions. A verified tag appears only where Hugging Face marks an entry as verified.
Original board: huggingface.co/datasets/internlm/WildClawBench
Part of the Hugging Face official benchmarks map by QuantID. Unofficial. Not affiliated with or endorsed by Hugging Face.