OpenEvals - EvasionBench: leaderboard
Hugging Face OpenEvals official leaderboard covering EvasionBench evasive-answer detection in financial text.
Metric: Accuracy (%). Source: huggingface.co. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-4.7 | 82.91 | #192 |
| 2 | Qwen3-Coder-480B-A35B-Instruct | 78.16 | |
| 3 | MiniMax-M2.1 | 71.31 | #289 |
| 4 | DeepSeek-V3.2 | 66.88 | |
| 5 | Kimi-K2-Instruct-0905 | 66.68 |
Interactive version: theaggregate.ai/benchmark?slug=openevals-evasionbench · How It Works · Data refreshed daily, snapshot 2026-10-20.