Up to four models
GPT-6.1 Sol vs Claude Fable 5.1
On the index (GPT-4 = 100), GPT-6.1 Sol scores 191 and Claude Fable 5.1 189. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-6.1 Sol
191175–207
WeightsClosed
ReleasedSep 2026
Price in / out$2.0 / $10
Context1.05M
Core tests taken6 of 9
Claude Fable 5.1
189173–205
WeightsClosed
ReleasedSep 2026
Price in / out$10 / $50
Context1M
Core tests taken6 of 9
GPT-6.1 Sol and Claude Fable 5.1 are too close to call (2 points apart, within the uncertainty), ahead on 2 of 5 shared core tests, and costs 5.0× less.
Test by testthird-party results
GPT-6.1 SolClaude Fable 5.1
Every result as a table
| Test | GPT-6.1 Sol | Claude Fable 5.1 |
|---|---|---|
| Reasoning & knowledge | ||
| LiveBench Data Analysis | 82.7% | 80.3% |
| LiveBench IF | 74.2% | 73.0% |
| LiveBench Language | 90.1% | 89.5% |
| LiveBench Reasoning | 92.6% | 91.7% |
| EBR-bench | 54.3% | 57.1% |
| SimpleBench | 82.9% | 86.6% |
| SimpleQA Verified | 73.9% | 70.8% |
| Maths | ||
| FrontierMath (tiers 1-3) | 93.7% | 90.2% |
| AIME-style maths (OTIS mock) | 100.0% | 100.0% |
| LiveBench Mathematics | 96.8% | 97.0% |
| ProofBench v1.1 | 99.0% | 100.0% |
| Chess puzzles | 68.0% | 48.0% |
| FrontierMath (tier 4) | 100.0% | 87.8% |
| Coding | ||
| Terminal-Bench 4.0 | 55.0% | 58.1% |
| LiveBench Agentic Coding | 56.8% | 66.1% |
| LiveBench Coding | 80.7% | 86.4% |
| Code Migration | 65.1% | 54.6% |
| CyberBench v1.1 | 39.3% | 70.4% |
| IOI | 96.9% | 90.8% |
| ProgramBench | 3.0% | 7.0% |
| Vibe Code Bench v1.1 | 88.9% | 90.3% |
| MirrorCode | 61.1% | 93.3% |
| Agents | ||
| APEX-Agents | 60.0% | 68.6% |
| SRE Bench | 50.8% | 22.9% |
| Terminal-Bench-Science | 52.9% | 40.0% |
| DTBench | 96.5% | 97.6% |
| Furniture assembly | 80.0% | 70.0% |
| LMCA | 62.8% | 65.5% |
| Mystery game puzzles | 80.0% | 58.0% |
| Professional work | ||
| Finance Agent (v2) | 52.0% | 58.9% |
| EMB | 70.8% | 76.7% |
| Legal Research Bench | 38.5% | 55.3% |
| MedCode | 48.8% | 53.5% |
| MedScribe | 86.5% | 91.3% |
| Public Benefits Bench v1.1 | 59.3% | 74.9% |
| Tax Agent Bench | 21.8% | 50.7% |
| Novel problems | ||
| ARC-AGI-2 | 94.2% | 90.0% |
| ARC-AGI-1 | 98.5% | 97.5% |
| MysteryMechanism | 46.4% | 47.8% |
Tests at least two of these models have taken. Bold rows are the core tests.