Model comparison
Claude Fable 5 vs GPT-5.5
Claude Fable 5 is the stronger model overall, scoring 66.8 to 63.4 on the Noometry Index. GPT-5.5 costs 1.8× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.
Last verified . 60 shared benchmarks.
Summary
- They share 60 benchmarks with published results for both. Claude Fable 5 scores higher in 7 categories and GPT-5.5 in 3 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in coding, where Claude Fable 5 leads 70.6 to 58.2.
- The biggest single-benchmark swing is MirrorCode: 63.9% for Claude Fable 5 and 10% for GPT-5.5.
- GPT-5.5 is cheaper at $5 / $30 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- GPT-5.5 accepts more context: 1.05M tokens versus 1M.
Side by side
| Claude Fable 5 | GPT-5.5 | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.8 | 63.4 |
| Released | 2026-06-07 | 2026-04-23 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1.05M |
| Max output | 128K | 128K |
| Input $ / M tokens | $10 | $5 |
| Output $ / M tokens | $50 | $30 |
| Results tracked | 62 | 71 |
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-5.5: 58.2 (#17)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| DeepSWE | 69.9% | 67% |
| FrontierCode | 53.5% | 43% |
| LMArena WebDev | 1625 | 1513 |
| SciCode | 61% | 56.1% |
| GSO | 78.4% | 40.2% |
| WeirdML | 91.9% | 84.9% |
| LMArena Coding | 1519 | 1494 |
| MirrorCode | 63.9% | 10% |
| ALE-Bench | 2,041 | 1,943 |
| SWE-bench Verified | — | 80.6% |
| FrontierSWE | 47% | — |
Agentic & Tool Use Claude Fable 5 leads
Claude Fable 5: 54.0 (#2), GPT-5.5: 50.7 (#6)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| APEX-Agents | 63.6% | 55.1% |
| Remote Labor Index | 16.1% | 6.3% |
| τ²-bench Banking | 39.7% | 44.6% |
| PostTrainBench | 41.8% | 27.2% |
| GBAEval | 74.5% | 53.2% |
| GDP.pdf | 30% | 26% |
| LMArena Search | 1230 | 1242 |
| Vending-Bench 2 | 5,680 | 7,524 |
| Terminal-Bench | — | 84.7% |
| OSWorld 2.0 | — | 13% |
| DeepResearch Bench | — | 54% |
| ExploitBench | — | 47.4% |
Reasoning Claude Fable 5 leads
Claude Fable 5: 76.8 (#6), GPT-5.5: 72.8 (#11)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| ARC-AGI-2 | 89.2% | 85% |
| SimpleBench | 81.9% | 69% |
| Kagi LLM Benchmark | 91.4% | 88.8% |
| NYT Connections (extended) | 92.7% | 96.2% |
| ARC-AGI-1 | 98.5% | 95% |
| CritPt | 28.6% | 27.1% |
| Chess Puzzles | 41% | 54% |
| EBR-Bench | 39.5% | 34.3% |
| LMArena Hard Prompts | 1508 | 1489 |
| Mystery Game Puzzles | 52% | 56% |
| DTBench | 98.4% | 96% |
| LMCA | 61.1% | 54.3% |
| Surface Evolver Bench | 95% | 88.1% |
| Bench to the Future 3 | 0.13 | 0.14 |
| Epoch Capabilities Index | 162.06 | 159.1 |
| EnigmaEval | 39.3% | — |
| ForecastBench | — | 60.6 |
Math Claude Fable 5 leads
Claude Fable 5: 88.5 (#5), GPT-5.5: 81.7 (#11)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 85.3% |
| FrontierMath Tier 4 | 90.2% | 72.5% |
| OTIS Mock AIME 2024-2025 | 100% | 100% |
| ProofBench | 95% | 50% |
| LMArena Math | 1519 | 1486 |
| FrontierMath Erdős | 0% | 0% |
| MathArena Final-Answer Competitions | — | 94.3% |
| FrontierMath (Feb 2025 set) | — | 51.7% |
| FrontierMath Tier 4 (v1) | — | 35.4% |
Knowledge GPT-5.5 leads
Claude Fable 5: 62.2 (#25), GPT-5.5: 64.4 (#17)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| GPQA Diamond | 85.9% | 94% |
| SimpleQA Verified | 70.7% | 63% |
| LMArena Expert | 1534 | 1508 |
| Vectara Hallucination Rate | — | 9.3% |
Multimodal GPT-5.5 leads
Claude Fable 5: 45.3 (#17), GPT-5.5: 46.9 (#12)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| LMArena Vision | 1324 | 1297 |
| Blueprint-Bench 2 | 38.6% | 36.2% |
| Furniture Assembly | 35.8% | 44.2% |
| LMArena Document | 1496 | 1486 |
Multilingual Too close to call
Claude Fable 5: 57.3 (#9), GPT-5.5: 56.4 (#20)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| LMArena Non-English | 1481 | 1467 |
| LMArena Chinese | 1543 | 1533 |
| LMArena French | 1505 | 1486 |
| LMArena German | 1486 | 1480 |
| LMArena Japanese | 1506 | 1498 |
| LMArena Korean | 1488 | 1460 |
| LMArena Russian | 1504 | 1473 |
| LMArena Spanish | 1498 | 1468 |
Instruction Following Claude Fable 5 leads
Claude Fable 5: 78.6 (#8), GPT-5.5: 77.5 (#18)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| LMArena Instruction Following | 1502 | 1479 |
Long Context GPT-5.5 leads
Claude Fable 5: 46.3 (#23), GPT-5.5: 48.3 (#12)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| LMArena Longer Query | 1509 | 1484 |
| CL-bench Life | — | 22.2% |
Writing & Preference Claude Fable 5 leads
Claude Fable 5: 75.9 (#5), GPT-5.5: 72.7 (#13)
| Benchmark | Claude Fable 5 | GPT-5.5 |
|---|---|---|
| LMArena Text | 1491 | 1472 |
| LMArena Creative Writing | 1494 | 1455 |
| EQ-Bench Creative Writing | 1943 | 1844 |
| EQ-Bench 4 | 1340 | 1315 |
| LMArena Multi-Turn | 1504 | 1476 |
Frequently asked questions
Is Claude Fable 5 better than GPT-5.5?
Claude Fable 5 is the stronger model overall, scoring 66.8 to 63.4 on the Noometry Index. GPT-5.5 costs 1.8× 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-5.5?
GPT-5.5 is cheaper. It lists at $5 per million input tokens and $30 per million output tokens; Claude Fable 5 lists at $10 and $50.
Is Claude Fable 5 or GPT-5.5 better for coding?
Claude Fable 5 scores higher on coding benchmarks: 70.6 versus 58.2 in the Noometry coding category.
Which has the bigger context window?
GPT-5.5 does, with 1.05M tokens against 1M.
How many benchmarks do Claude Fable 5 and GPT-5.5 share?
60 benchmarks have published results for both models. Claude Fable 5 has 62 scored results on Noometry and GPT-5.5 has 71.