MintEval (200-Task Subset): leaderboard
Metric: ActionMatch (%), open setting, compile failures scored 0, 200 stratified BTCUSDT tasks. Source: github.com. Saturation forecast: Around January 2027. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5.5 | 88.9 |
| 2 | DeepSeek V4 Pro (High) | 71.2 |
| 3 | GPT-5.4 Mini | 54.3 |
Interactive version: theaggregate.ai/benchmark?slug=minteval-200-task-subset · How It Works · Data refreshed daily, snapshot 2026-10-06.