Claude Fable 5.1 scores 69.4 on the Respan Index (#4), ahead of GPT-5.6 Terra at 53.5 (#14). Claude Fable 5.1 leads on 21 of the 23 benchmarks both report. GPT-5.6 Terra is 4.4x cheaper per token at list price. GPT-5.6 Terra has the larger context window (1.1M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| Furniture Assembly (Epoch AI) |
| 70% |
| 54.2% |
| LiveBench Agentic Coding (LiveBench) | 66.1% | 54.9% |
| LiveBench Coding (LiveBench) | 86.4% | 78.2% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 79.3% |
| LiveBench Instruction Following (LiveBench) | 73% | 64.6% |
| LiveBench Language (LiveBench) | 89.5% | 82.9% |
| LiveBench Mathematics (LiveBench) | 97% | 94.9% |
| LiveBench Reasoning (LiveBench) | 91.7% | 90.6% |
| Mystery Game Puzzles (Epoch AI) | 58% | 35% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 99.7% |
| SAGE (Vals AI) | 48.5% | 47% |
| SimpleQA Verified | 70.8% | 43.2% |
| Terminal-Bench 4.0 (Vals AI) | 58.1% | 22.7% |
| Vending-Bench 2 (Andon Labs) | 5421.56 | 7343.21 |
| WeirdML (Håvard Tveit Ihle) | 92.9% | 78.3% |
| LMArena Agent (LMArena) | 0.1431 | 0.004 |
| LMArena Elo | 1501 | 1466 |
| LMArena Vision (LMArena) | 1288 | 1268 |
| SimpleBench (SimpleBench) | 86.6% | 48.9% |
Higher is better. Each score links to where it was published.