DeepSeek V4.1 Flash-1017 14.18
DeepSeek · DSH · 1M
- Score
- 14.18
- Rank
- 7/ 10
- Best on
- 0/ 8 tasks
- API cost
- $20.0
- Input tokens · 99.6% cached
- 2025 M
- Output tokens
- 4.7 M
§ 01Per task
| Task | Judged | Rank | Tier | Self-best | Rounds | Hours | API cost | |
|---|---|---|---|---|---|---|---|---|
| LLaVA | 2.22× | 7 / 10 | N1 | 2.23× | 32 | 7.7 h | $2.58 | trace → |
| YOCO | 1.55× | 9 / 10 | N1 | 1.68× | 74 | 7.9 h | $3.16 | trace → |
| NanoChat | 1.06× | 8 / 10 | N1 | 1.06× | 87 | 9.0 h | $2.74 | trace → |
| BAGEL SFT | 1.15× | 10 / 10 | N0 | 1.23× | 22 | 10.0 h | $2.06 | trace → |
| R3 | 1.19× | 10 / 10 | N0 | 1.28× | 16 | 10.0 h | $1.93 | trace → |
| 3DGS | 1.71× | 9 / 10 | N1 | 1.72× | 52 | 9.7 h | $4.06 | trace → |
| Slime GRPO | 1.88× | 5 / 10 | N1 | 1.94× | 10 | 10.0 h | $1.41 | trace → |
| Slime OPD | 1.43× | 6 / 10 | N1 | 1.43× | 13 | 10.0 h | $2.06 | trace → |