Claude Fable 5.1 scores 69.4 on the Respan Index (#4), ahead of DeepSeek V4.1 Flash at 50.9 (#20). Claude Fable 5.1 leads on 13 of the 14 benchmarks both report. DeepSeek V4.1 Flash is 38x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| LiveBench Instruction Following (LiveBench) | 73% | 70% |
| LiveBench Language (LiveBench) | 89.5% | 81.2% |
| LiveBench Mathematics (LiveBench) | 97% | 93.3% |
| LiveBench Reasoning (LiveBench) | 91.7% | 86.7% |
| SAGE (Vals AI) | 48.5% | 47.9% |
| Terminal-Bench 4.0 (Vals AI) | 58.1% | 19.7% |
| LMArena Agent (LMArena) | 0.1431 | 0.0402 |
| LMArena Elo | 1501 | 1474 |
| LMArena WebDev (LMArena) | 1749 | 1620 |
| SimpleBench (SimpleBench) | 86.6% | 66.7% |
Higher is better. Each score links to where it was published.