DeepSeek V4.1 Flash scores 50.9 on the Respan Index (#20), ahead of Kimi K2.7 Code at 34.1 (#58). DeepSeek V4.1 Flash leads on 11 of the 11 benchmarks both report. DeepSeek V4.1 Flash is 3.3x cheaper per token at list price. DeepSeek V4.1 Flash has the larger context window (1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| LiveBench Instruction Following (LiveBench) | 70% | 56.3% |
| LiveBench Language (LiveBench) | 81.2% | 77.9% |
| LiveBench Mathematics (LiveBench) | 93.3% | 79.6% |
| LiveBench Reasoning (LiveBench) | 86.7% | 82.8% |
| LMArena WebDev (LMArena) | 1620 | 1473 |
| SimpleBench (SimpleBench) | 66.7% | 57.9% |
| GPQA Diamond | 90.9% | 87.9% |
Higher is better. Each score links to where it was published.