Claude Fable 5.1 scores 69.4 on the Respan Index (#4), ahead of GPT-5.6 Sol at 60.6 (#10). Claude Fable 5.1 leads on 21 of the 27 benchmarks both report. GPT-5.6 Sol is 2.5x cheaper per token at list price. GPT-5.6 Sol has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| FrontierMath Tiers 1-3 v2 (Epoch AI) |
| 90.2% |
| 89.1% |
| FrontierSWE V2 (Proximal Labs) | 56.3% | 32.2% |
| Furniture Assembly (Epoch AI) | 70% | 56.7% |
| LiveBench Agentic Coding (LiveBench) | 66.1% | 56.2% |
| LiveBench Coding (LiveBench) | 86.4% | 83.9% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 79.8% |
| LiveBench Instruction Following (LiveBench) | 73% | 71.8% |
| LiveBench Language (LiveBench) | 89.5% | 87.7% |
| LiveBench Mathematics (LiveBench) | 97% | 96.2% |
| LiveBench Reasoning (LiveBench) | 91.7% | 91.7% |
| MirrorCode (Epoch AI) | 73.3% | 20% |
| Mystery Game Puzzles (Epoch AI) | 58% | 58% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 100% |
| SAGE (Vals AI) | 48.5% | 52.6% |
| SimpleQA Verified | 70.8% | 69.7% |
| Terminal-Bench 4.0 (Vals AI) | 58.1% | 37.9% |
| Vending-Bench 2 (Andon Labs) | 5421.56 | 9619.37 |
| WeirdML (Håvard Tveit Ihle) | 92.9% | 89.4% |
| LMArena Agent (LMArena) | 0.1431 | 0.0645 |
| LMArena Elo | 1501 | 1484 |
| LMArena Vision (LMArena) | 1288 | 1285 |
| GSO Opt@1 (GSO) | 87.3% | 70.6% |
| SimpleBench (SimpleBench) | 86.6% | 71.7% |
Higher is better. Each score links to where it was published.