GPT-6 Astrasuspected nerf 13.41
OpenAI · Codex · 272k
- Score
- 13.41
- Rank
- 9/ 10
- Best on
- 0/ 8 tasks
- API cost
- $1158
- Input tokens · 97.4% cached
- 776 M
- Output tokens
- 3.4 M
§ 01Per task
| Task | Judged | Rank | Tier | Self-best | Rounds | Hours | API cost | |
|---|---|---|---|---|---|---|---|---|
| LLaVA | 1.41× | 9 / 10 | N1 | 1.29× | 40 | 10.0 h | $187 | trace → |
| YOCO | 1.79× | 5 / 10 | N1 | 1.79× | 25 | 10.0 h | $99.7 | trace → |
| NanoChat | 1.13× | 4 / 10 | N1 | 1.21× | 27 | 10.0 h | $143 | trace → |
| BAGEL SFT | 1.62× | 5 / 10 | N1 | 1.65× | 15 | 10.0 h | $78.9 | trace → |
| R3 | 1.47× | 7 / 10 | N0 | 1.54× | 22 | 10.0 h | $89.3 | trace → |
| 3DGS | 2.04× | 7 / 10 | N1 | 2.07× | 22 | 10.0 h | $97.5 | trace → |
| Slime GRPO | 1.15× | 10 / 10 | N1 | 1.13× | 20 | 10.0 h | $217 | trace → |
| Slime OPD | 1.30× | 9 / 10 | N1 | 1.38× | 22 | 10.0 h | $245 | trace → |