Official benchmarks map / EvasionBench
zai-org/GLM-4.7 is ranked first on the EvasionBench leaderboard with 82.91, 6.1% ahead of Qwen/Qwen3-Coder-480B-A35B-Instruct. Read live from the Hugging Face API on November 6, 2026.
| # | Model | Organization | Score |
|---|---|---|---|
| 1 | zai-org/GLM-4.7 | zai-org | 82.91 |
| 2 | Qwen/Qwen3-Coder-480B-A35B-Instruct | Qwen | 78.16 |
| 3 | FINAL-Bench/Darwin-180B-RSI-R3 | FINAL-Bench | 77.83 |
| 4 | MiniMaxAI/MiniMax-M2.1 | MiniMaxAI | 71.31 |
| 5 | moonshotai/Kimi-K2-Instruct-0905 | moonshotai | 66.68 |
EvasionBench 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/FutureMa/EvasionBench
Part of the Hugging Face official benchmarks map by QuantID. Unofficial. Not affiliated with or endorsed by Hugging Face.