Model comparison
Claude Fable 5 vs GPT-5.1
Claude Fable 5 is the stronger model overall, scoring 66.8 to 49.0 on the Noometry Index. GPT-5.1 costs 5.8× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Last verified . 39 shared benchmarks.
Summary
- They share 39 benchmarks with published results for both. Claude Fable 5 scores higher in 8 categories and GPT-5.1 in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5 leads 76.8 to 39.8.
- The biggest single-benchmark swing is ARC-AGI-2: 89.2% for Claude Fable 5 and 17.6% for GPT-5.1.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- Claude Fable 5 accepts more context: 1M tokens versus 400K.
Side by side
| Claude Fable 5 | GPT-5.1 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.8 | 49.0 |
| Released | 2026-06-07 | 2025-11-13 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 400K |
| Max output | 128K | 128K |
| Input $ / M tokens | $10 | $1.25 |
| Output $ / M tokens | $50 | $10 |
| Results tracked | 62 | 63 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5 leads
Claude Fable 5: 70.6 (#4), GPT-5.1: 46.4 (#66)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena WebDev | 1625 | 1395 |
| SciCode | 61% | 43.3% |
| GSO | 78.4% | 13.7% |
| WeirdML | 91.9% | 60.8% |
| LMArena Coding | 1519 | 1454 |
| ALE-Bench | 2,041 | 1,192 |
| SWE-bench Verified | — | 68% |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| SWE-bench Verified (bash only) | — | 66% |
| FrontierSWE | 47% | — |
| LiveBench Coding | — | 72.5% |
| MirrorCode | 63.9% | — |
Agentic & Tool Use Claude Fable 5 leads
Claude Fable 5: 54.0 (#2), GPT-5.1: 32.7 (#60)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Search | 1230 | 1199 |
| Vending-Bench 2 | 5,680 | 1,473 |
| Terminal-Bench | — | 47.6% |
| APEX-Agents | 63.6% | — |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| DeepResearch Bench | — | 42.8% |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), GPT-5.1: 39.8 (#58)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 89.2% | 17.6% |
| SimpleBench | 81.9% | 53.2% |
| ARC-AGI-1 | 98.5% | 72.8% |
| CritPt | 28.6% | 4.9% |
| Chess Puzzles | 41% | 32% |
| EnigmaEval | 39.3% | 11.2% |
| LMArena Hard Prompts | 1508 | 1457 |
| Mystery Game Puzzles | 52% | 19% |
| DTBench | 98.4% | 90.1% |
| LMCA | 61.1% | 43.9% |
| Epoch Capabilities Index | 162.06 | 149.64 |
| Kagi LLM Benchmark | 91.4% | — |
| NYT Connections (extended) | 92.7% | — |
| EBR-Bench | 39.5% | — |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| ForecastBench | — | 58.1 |
| LiveBench | — | 78.8% |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), GPT-5.1: 52.2 (#51)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 100% | 88.6% |
| LMArena Math | 1519 | 1447 |
| FrontierMath (Tiers 1-3) | 87% | — |
| FrontierMath Tier 4 | 90.2% | — |
| ProofBench | 95% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 12.5% |
Knowledge Claude Fable 5 leads
Claude Fable 5: 62.2 (#25), GPT-5.1: 50.6 (#71)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 85.9% | 87.6% |
| SimpleQA Verified | 70.7% | 48% |
| LMArena Expert | 1534 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| MMLU-Pro | — | 57.9% |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |
Multimodal Too close to call
Claude Fable 5: 45.3 (#17), GPT-5.1: 44.8 (#19)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1324 | 1250 |
| LMArena Document | 1496 | 1403 |
| VPCT | — | 58.7% |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
Multilingual Claude Fable 5 leads
Claude Fable 5: 57.3 (#9), GPT-5.1: 53.8 (#56)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1481 | 1431 |
| LMArena Chinese | 1543 | 1495 |
| LMArena French | 1505 | 1450 |
| LMArena German | 1486 | 1438 |
| LMArena Japanese | 1506 | 1453 |
| LMArena Korean | 1488 | 1401 |
| LMArena Russian | 1504 | 1435 |
| LMArena Spanish | 1498 | 1433 |
Instruction Following GPT-5.1 leads
Claude Fable 5: 78.6 (#8), GPT-5.1: 83.9 (#1)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1502 | 1443 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |
Long Context GPT-5.1 leads
Claude Fable 5: 46.3 (#23), GPT-5.1: 47.6 (#14)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1509 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), GPT-5.1: 64.5 (#55)
| Benchmark | Claude Fable 5 | GPT-5.1 |
|---|---|---|
| LMArena Text | 1491 | 1443 |
| LMArena Creative Writing | 1494 | 1427 |
| LMArena Multi-Turn | 1504 | 1450 |
| EQ-Bench Creative Writing | 1943 | — |
| WildBench | — | 86.3% |
| EQ-Bench 4 | 1340 | — |
| LiveBench Language | — | 80.2% |
Frequently asked questions
Is Claude Fable 5 better than GPT-5.1?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 49.0 on the Noometry Index. GPT-5.1 costs 5.8× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Which is cheaper, Claude Fable 5 or GPT-5.1?
GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Fable 5 lists at $10 and $50.
Is Claude Fable 5 or GPT-5.1 better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 46.4 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5 does, with 1M tokens against 400K.
How many benchmarks do Claude Fable 5 and GPT-5.1 share?
39 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and GPT-5.1 has 63.