Grok 4.6 scores 51.0 on the Respan Index (#19), ahead of DeepSeek V4.1 Flash at 50.9 (#20). DeepSeek V4.1 Flash leads on 10 of the 16 benchmarks both report. DeepSeek V4.1 Flash is 5.7x 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% |
| 71.9% |
| LiveBench Language (LiveBench) | 81.2% | 83.7% |
| LiveBench Mathematics (LiveBench) | 93.3% | 92.6% |
| LiveBench Reasoning (LiveBench) | 86.7% | 90.5% |
| SAGE (Vals AI) | 47.9% | 28.9% |
| Terminal-Bench 4.0 (Vals AI) | 19.7% | 17.2% |
| LMArena Agent (LMArena) | 0.0402 | 0.0128 |
| LMArena Elo | 1474 | 1454 |
| LMArena WebDev (LMArena) | 1620 | 1620 |
| SimpleBench (SimpleBench) | 66.7% | 75.9% |
| DeepSWE v1.1 | 74.2% | 65.9% |
| GPQA Diamond | 90.9% | 94% |
Higher is better. Each score links to where it was published.