Claude Opus 4.8 scores 52.2 on the Respan Index (#16), ahead of Grok 4.6 at 51.0 (#19). Grok 4.6 leads on 14 of the 27 benchmarks both report. Grok 4.6 is 3.3x cheaper per token at list price. Claude Opus 4.8 has the larger context window (1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| 56.1% |
| 31.7% |
| FrontierMath Tiers 1-3 v2 (Epoch AI) | 80% | 66% |
| Furniture Assembly (Epoch AI) | 42.5% | 40% |
| GPQA Diamond | 93.6% | 94% |
| LiveBench Agentic Coding (LiveBench) | 50.5% | 57% |
| LiveBench Coding (LiveBench) | 81.8% | 76.8% |
| LiveBench Data Analysis (LiveBench) | 66% | 73.9% |
| LiveBench Instruction Following (LiveBench) | 72% | 71.9% |
| LiveBench Language (LiveBench) | 79.7% | 83.7% |
| LiveBench Mathematics (LiveBench) | 94.3% | 92.6% |
| LiveBench Reasoning (LiveBench) | 89.2% | 90.5% |
| Mystery Game Puzzles (Epoch AI) | 36% | 34% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 98.3% | 99.2% |
| SAGE (Vals AI) | 54.8% | 28.9% |
| SimpleQA Verified | 53% | 49.3% |
| Terminal-Bench 4.0 (Vals AI) | 23.2% | 17.2% |
| Vending-Bench 2 (Andon Labs) | 5787.43 | 9047.03 |
| WeirdML (Håvard Tveit Ihle) | 82.9% | 67.3% |
| LMArena Agent (LMArena) | 0.0664 | 0.0128 |
| LMArena Elo | 1453 | 1454 |
| LMArena Vision (LMArena) | 1286 | 1264 |
| LMArena WebDev (LMArena) | 1556 | 1620 |
| SimpleBench (SimpleBench) | 64.8% | 75.9% |
Higher is better. Each score links to where it was published.