Claude Fable 5.1 scores 69.4 on the Respan Index (#4), ahead of Gemini 3.8 Flash at 52.2 (#16). Claude Fable 5.1 leads on 22 of the 25 benchmarks both report. Gemini 3.8 Flash is 13x cheaper per token at list price. Gemini 3.8 Flash has the larger context window (1.0M tokens).
* Estimated: no published scores in that category. How the Respan Index works
| FrontierSWE V2 (Proximal Labs) |
| 56.3% |
| 19.6% |
| Furniture Assembly (Epoch AI) | 70% | 31.7% |
| LiveBench Agentic Coding (LiveBench) | 66.1% | 54.2% |
| LiveBench Coding (LiveBench) | 86.4% | 72.5% |
| LiveBench Data Analysis (LiveBench) | 80.3% | 54% |
| LiveBench Instruction Following (LiveBench) | 73% | 81.4% |
| LiveBench Language (LiveBench) | 89.5% | 87.8% |
| LiveBench Mathematics (LiveBench) | 97% | 91.6% |
| LiveBench Reasoning (LiveBench) | 91.7% | 89.3% |
| Mystery Game Puzzles (Epoch AI) | 58% | 47% |
| OTIS Mock AIME 2024-2025 (Epoch AI) | 100% | 98.9% |
| SAGE (Vals AI) | 48.5% | 35.1% |
| SimpleQA Verified | 70.8% | 69.7% |
| Terminal-Bench 4.0 (Vals AI) | 58.1% | 19.2% |
| Vending-Bench 2 (Andon Labs) | 5421.56 | 5093.79 |
| WeirdML (Håvard Tveit Ihle) | 92.9% | 84.8% |
| LMArena Agent (LMArena) | 0.1431 | 0.0297 |
| LMArena Elo | 1501 | 1495 |
| LMArena Vision (LMArena) | 1288 | 1290 |
| LMArena WebDev (LMArena) | 1749 | 1583 |
| SimpleBench (SimpleBench) | 86.6% | 82.4% |
Higher is better. Each score links to where it was published.