Claude Opus 5.5 40.00
Anthropic · Claude Code · 1M
- Score
- 40.00
- Rank
- 1/ 10
- Best on
- 4/ 8 tasks
- API cost
- $644
- Input tokens · 99.4% cached
- 1931 M
- Output tokens
- 8.0 M
§ 01Per task
| Task | Judged | Rank | Tier | Self-best | Rounds | Hours | API cost | |
|---|---|---|---|---|---|---|---|---|
| LLaVA | 2.92× | 1 / 10 | N1 | 2.77× | 37 | 10.0 h | $90.4 | trace → |
| YOCO | 2.68× | 1 / 10 | N1 | 2.68× | 18 | 10.0 h | $57.6 | trace → |
| NanoChat | 1.20× | 2 / 10 | N1 | 1.10× | 5 | 10.0 h | $78.1 | trace → |
| BAGEL SFT | 2.20× | 2 / 10 | N1 | 2.22× | 8 | 9.3 h | $48.9 | trace → |
| R3 | 2.41× | 1 / 10 | N0 | 2.47× | 8 | 10.0 h | $129 | trace → |
| 3DGS | 5.82× | 1 / 10 | N1 | 5.85× | 20 | 10.0 h | $65.4 | trace → |
| Slime GRPO | 2.35× | 2 / 10 | N1 | 2.48× | 7 | 10.0 h | $87.7 | trace → |
| Slime OPD | 1.64× | 2 / 10 | N1 | 1.63× | 13 | 10.0 h | $86.8 | trace → |