Claude Fable 5.1 scores 69.4 on the Respan Index (#4), ahead of GPT-5.5 at 52.5 (#15). Claude Fable 5.1 leads on 22 of the 25 benchmarks both report. GPT-5.5 is 1.8x cheaper per token at list price.
* Estimated: no published scores in that category. How the Respan Index works
| 90.2% |
| 85.3% |
| Furniture Assembly (Epoch AI) | 70% | 44.2% |
| LiveBench Agentic Coding (LiveBench) | 66.1% | 54% |
| LiveBench Coding (LiveBench) | 86.4% | 82.1% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 81.6% |
| LiveBench Instruction Following (LiveBench) | 73% | 70.7% |
| LiveBench Language (LiveBench) | 89.5% | 87.4% |
| LiveBench Mathematics (LiveBench) | 97% | 95.9% |
| LiveBench Reasoning (LiveBench) | 91.7% | 89.7% |
| MirrorCode (Epoch AI) | 73.3% | 10% |
| Mystery Game Puzzles (Epoch AI) | 58% | 56% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 84.4% |
| SAGE (Vals AI) | 48.5% | 51.5% |
| SimpleQA Verified | 70.8% | 63% |
| Vending-Bench 2 (Andon Labs) | 5421.56 | 7523.84 |
| WeirdML (Håvard Tveit Ihle) | 92.9% | 84.9% |
| LMArena Agent (LMArena) | 0.1431 | 0.0415 |
| LMArena Elo | 1501 | 1466 |
| LMArena Vision (LMArena) | 1288 | 1287 |
| GSO Opt@1 (GSO) | 87.3% | 37.3% |
| SimpleBench (SimpleBench) | 86.6% | 69% |
Higher is better. Each score links to where it was published.