# Claude Fable 5.1 vs GPT-5

> Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 50.9 on the Noometry Index. GPT-5 costs 5.8× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-fable-5-1-vs-gpt-5
- Last updated: 2026-10-10
- Shared benchmarks: 40

## Summary

- They share 40 benchmarks with published results for both. Claude Fable 5.1 scores higher in 9 categories and GPT-5 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Fable 5.1 leads 76.7 to 38.3.
- The biggest single-benchmark swing is ProofBench: 100% for Claude Fable 5.1 and 18% for GPT-5.
- GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Claude Fable 5.1 accepts more context: 1M tokens versus 400K.

## Snapshot

| | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 69.0 | 50.9 |
| Rank | 2 | 45 |
| Context | 1M | 400K |
| Input $/M | $10 | $1.25 |
| Output $/M | $50 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Fable 5.1: 74.7 (#1)
- GPT-5: 50.3 (#47)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena WebDev | 1744 | 1418 |
| SciCode | 63.1% | 42.9% |
| GSO | 88.2% | 6.9% |
| WeirdML | 92.9% | 60.7% |
| LMArena Coding | 1528 | 1436 |
| ALE-Bench | 2,143 | 1,162 |
| SWE-bench Verified | — | 73.6% |
| FrontierCode | 50.9% | — |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| CursorBench | 51.8% | — |
| FrontierSWE | 56.3% | — |
| MirrorCode | 73.3% | — |
| AlgoTune | — | 1.67 |

## Agentic & Tool Use

- Claude Fable 5.1: 50.7 (#5)
- GPT-5: 33.1 (#56)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| Remote Labor Index | 17.9% | 1.7% |
| Terminal-Bench | — | 49.6% |
| APEX-Agents | 68.6% | — |
| GDPval | — | 34.8% |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| GDP.pdf | 29.6% | — |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |
| Vending-Bench 2 | 5,422 | — |

## Reasoning

- Claude Fable 5.1: 76.7 (#7)
- GPT-5: 38.3 (#64)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| ARC-AGI-2 | 90% | 9.9% |
| ARC-AGI-1 | 97.5% | 65.7% |
| CritPt | 31.1% | 12.6% |
| Chess Puzzles | 47% | 37% |
| EBR-Bench | 57.1% | 12.7% |
| LMArena Hard Prompts | 1526 | 1416 |
| Mystery Game Puzzles | 58% | 23% |
| DTBench | 97.6% | 90.7% |
| LMCA | 65.5% | 40% |
| Epoch Capabilities Index | 164.7 | 150 |
| SimpleBench | — | 56.7% |
| Kagi LLM Benchmark | — | 72.7% |
| NYT Connections (extended) | 90% | — |
| EnigmaEval | — | 10.5% |
| ForecastBench | — | 61.4 |

## Math

- Claude Fable 5.1: 89.6 (#4)
- GPT-5: 55.0 (#44)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 90.2% | 55.4% |
| FrontierMath Tier 4 | 87.8% | 22% |
| OTIS Mock AIME 2024-2025 | 100% | 91.4% |
| ProofBench | 100% | 18% |
| LMArena Math | 1525 | 1407 |
| Omni-MATH | — | 64.7% |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 12.5% |

## Knowledge

- Claude Fable 5.1: 69.6 (#6)
- GPT-5: 56.6 (#43)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| Humanity's Last Exam | 46.5% | 25.3% |
| SimpleQA Verified | 70.8% | 50.1% |
| LMArena Expert | 1535 | 1419 |
| GPQA Diamond | — | 86.2% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.3% |
| Vectara Hallucination Rate | — | 14.7% |
| GPQA (HELM) | — | 79.2% |

## Multimodal

- Claude Fable 5.1: 53.9 (#4)
- GPT-5: 46.8 (#13)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena Vision | 1318 | 1232 |
| GeoBench | — | 81% |
| VPCT | — | 66% |
| Blueprint-Bench 2 | 41.9% | — |
| Furniture Assembly | 70% | — |
| LMArena Document | 1513 | — |

## Multilingual

- Claude Fable 5.1: 59.1 (#3)
- GPT-5: 51.4 (#110)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena Non-English | 1507 | 1397 |
| LMArena Chinese | 1586 | 1422 |
| LMArena French | 1525 | 1410 |
| LMArena German | 1500 | 1416 |
| LMArena Japanese | 1543 | 1409 |
| LMArena Korean | 1534 | 1360 |
| LMArena Russian | 1521 | 1406 |
| LMArena Spanish | 1516 | 1399 |

## Instruction Following

- Claude Fable 5.1: 79.2 (#6)
- GPT-5: 73.8 (#113)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1388 |
| IFEval | — | 87.5% |

## Long Context

- Claude Fable 5.1: 46.7 (#20)
- GPT-5: 69.5 (#2)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1522 | 1399 |
| Fiction.LiveBench | — | 97.2% |

## Writing & Preference

- Claude Fable 5.1: 79.2 (#2)
- GPT-5: 63.4 (#65)

| Benchmark | Claude Fable 5.1 | GPT-5 |
|---|---|---|
| LMArena Text | 1510 | 1406 |
| LMArena Creative Writing | 1507 | 1365 |
| EQ-Bench Creative Writing | 2162 | 1627 |
| LMArena Multi-Turn | 1492 | 1426 |
| Short-Story Creative Writing | — | 86% |
| WildBench | — | 85.7% |

## FAQ

### Is Claude Fable 5.1 better than GPT-5?

Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 50.9 on the Noometry Index. GPT-5 costs 5.8× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.

### Which is cheaper, Claude Fable 5.1 or GPT-5?

GPT-5 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Fable 5.1 lists at $10 and $50.

### Is Claude Fable 5.1 or GPT-5 better for coding?

Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 50.3 in the Noometry coding category.

### Which has the bigger context window?

Claude Fable 5.1 does, with 1M tokens against 400K.

### How many benchmarks do Claude Fable 5.1 and GPT-5 share?

40 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and GPT-5 has 69.