Grok 4.7 scores 49.8 on the Respan Index (#23), ahead of Kimi K2.6 at 34.4 (#55). Grok 4.7 leads on 13 of the 19 benchmarks both report. Kimi K2.6 is 1.8x cheaper per token at list price. Grok 4.7 has the larger context window (500K tokens).
* Estimated: no published scores in that category. How the Respan Index works
| GPQA Diamond |
| 92.7% |
| 90.8% |
| LiveBench Agentic Coding (LiveBench) | 54% | 46.9% |
| LiveBench Coding (LiveBench) | 77.2% | 78.6% |
| LiveBench Data Analysis (LiveBench) | 76.9% | 65.1% |
| LiveBench Instruction Following (LiveBench) | 75.3% | 64.4% |
| LiveBench Language (LiveBench) | 80.1% | 75.1% |
| LiveBench Mathematics (LiveBench) | 95.7% | 84.3% |
| LiveBench Reasoning (LiveBench) | 82.7% | 79.4% |
| Mystery Game Puzzles (Epoch AI) | 29% | 18% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.1% | 96.1% |
| SAGE (Vals AI) | 40.8% | 50.2% |
| SimpleQA Verified | 56% | 34.9% |
| Vending-Bench 2 (Andon Labs) | 10536.83 | 6204.57 |
| LMArena Elo | 1443 | 1455 |
| LMArena WebDev (LMArena) | 1638 | 1509 |
Higher is better. Each score links to where it was published.