Tool Use on VitaBench
55.3Overall ScoreClaude-Opus-4.5
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude-Opus-4.52026.06 | 55.3 | |
| Gemini-3-pro High2026.06 | 48.8 | |
| Seed2.0 Pro2026.06 | 47 | |
| Gemini-3-Flash High2026.06 | 46.7 | |
| GPT-5.2 High2026.06 | 41.8 | |
| Seed2.0 Lite2026.06 | 41.8 | |
| Claude-Sonnet-4.52026.06 | 40.8 | |
| GPT-5-mini High2026.06 | 20.8 |