Up to four models
Claude Fable 5.2 vs GPT-6.1 Sol
On the index (GPT-4 = 100), Claude Fable 5.2 scores 202 and GPT-6.1 Sol 191. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
Claude Fable 5.2
202185–218
WeightsClosed
ReleasedNov 2026
Price in / out$8.0 / $40
Context1M
Core tests taken4 of 9
GPT-6.1 Sol
191175–207
WeightsClosed
ReleasedSep 2026
Price in / out$2.0 / $10
Context1.05M
Core tests taken6 of 9
Claude Fable 5.2 leads GPT-6.1 Sol by 11 points, ahead on 4 of 4 shared core tests, at 4.0× the price.
Test by testthird-party results
Claude Fable 5.2GPT-6.1 Sol
Every result as a table
| Test | Claude Fable 5.2 | GPT-6.1 Sol |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 96.0% | 95.4% |
| EBR-bench | 90.5% | 54.3% |
| SimpleBench | 94.9% | 82.9% |
| SimpleQA Verified | 94.0% | 73.9% |
| Maths | ||
| FrontierMath (tiers 1-3) | 95.1% | 93.7% |
| AIME-style maths (OTIS mock) | 99.7% | 100.0% |
| ProofBench v1.1 | 100.0% | 99.0% |
| Chess puzzles | 79.0% | 68.0% |
| FrontierMath (tier 4) | 100.0% | 100.0% |
| Coding | ||
| Terminal-Bench 4.0 | 65.1% | 55.0% |
| ProgramBench | 19.0% | 3.0% |
| Vibe Code Bench v1.1 | 93.8% | 88.9% |
| Agents | ||
| APEX-Agents | 80.1% | 60.0% |
| BioMysteryBench | 83.7% | 79.6% |
| Terminal-Bench-Science | 55.7% | 52.9% |
| DTBench | 99.2% | 96.5% |
| Furniture assembly | 90.0% | 80.0% |
| LMCA | 70.6% | 62.8% |
| Mystery game puzzles | 90.0% | 80.0% |
| Professional work | ||
| HLAB-Vals-1.1 | 5.0% | 6.3% |
| Legal Research Bench | 46.2% | 38.5% |
| Tax Agent Bench | 32.1% | 21.8% |
| Novel problems | ||
| MysteryMechanism | 51.8% | 46.4% |
Tests at least two of these models have taken. Bold rows are the core tests.