# Claude Fable 5 vs GPT-4.1

> Claude Fable 5 is the stronger model overall, scoring 66.3 to 35.7 on the Noometry Index. GPT-4.1 costs 5.7× less per token, which makes it the better buy when Claude Fable 5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-fable-5-vs-gpt-4-1
- Last updated: 2026-10-10
- Shared benchmarks: 34

## Summary

- They share 34 benchmarks with published results for both. Claude Fable 5 scores higher in 10 categories and GPT-4.1 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5 leads 88.4 to 21.4.
- The biggest single-benchmark swing is ARC-AGI-1: 98.5% for Claude Fable 5 and 5.5% for GPT-4.1.
- GPT-4.1 is cheaper at $2 / $8 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- GPT-4.1 accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.3 | 35.7 |
| Rank | 5 | 215 |
| Context | 1M | 1.05M |
| Input $/M | $10 | $2 |
| Output $/M | $50 | $8 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Fable 5: 67.0 (#4)
- GPT-4.1: 34.2 (#228)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| WeirdML | 91.9% | 39% |
| LMArena Coding | 1519 | 1391 |
| ALE-Bench | 2,041 | 558.1 |
| SWE-bench Verified | — | 48.5% |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| SWE-bench Verified (bash only) | — | 39.6% |
| Aider Polyglot | — | 52.4% |
| LMArena WebDev | 1626 | — |
| FrontierSWE | 47% | — |
| SciCode | 61% | — |
| MirrorCode | 63.9% | — |
| CadEval | — | 42% |

## Agentic & Tool Use

- Claude Fable 5: 54.1 (#2)
- GPT-4.1: 34.4 (#40)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| APEX-Agents | 63.6% | — |
| Berkeley Function Calling Leaderboard | — | 54% |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| PostTrainBench | 41.8% | — |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
| LMArena Search | 1230 | — |
| Vending-Bench 2 | 5,680 | — |

## Reasoning

- Claude Fable 5: 76.8 (#4)
- GPT-4.1: 11.7 (#334)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| ARC-AGI-2 | 89.2% | 0.4% |
| SimpleBench | 81.9% | 27% |
| Kagi LLM Benchmark | 91.4% | 52.3% |
| ARC-AGI-1 | 98.5% | 5.5% |
| Chess Puzzles | 41% | 6% |
| EnigmaEval | 39.3% | 2.2% |
| LMArena Hard Prompts | 1508 | 1384 |
| DTBench | 98.4% | 68.3% |
| LMCA | 61.1% | 25.6% |
| Epoch Capabilities Index | 163.6 | 136.79 |
| NYT Connections (extended) | 92.7% | — |
| CritPt | 28.6% | — |
| EBR-Bench | 39.5% | — |
| Mystery Game Puzzles | 52% | — |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| ForecastBench | — | 61.5 |

## Math

- Claude Fable 5: 88.4 (#5)
- GPT-4.1: 21.4 (#269)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 6% |
| OTIS Mock AIME 2024-2025 | 100% | 38.3% |
| LMArena Math | 1519 | 1369 |
| FrontierMath Tier 4 | 90.2% | — |
| ProofBench | 95% | — |
| Omni-MATH | — | 47.1% |
| MATH Level 5 | — | 83% |
| FrontierMath (Feb 2025 set) | — | 5.5% |
| FrontierMath Erdős | 0% | — |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Claude Fable 5: 62.3 (#22)
- GPT-4.1: 37.0 (#152)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| GPQA Diamond | 85.9% | 66.9% |
| SimpleQA Verified | 70.7% | 31.1% |
| LMArena Expert | 1534 | 1364 |
| Humanity's Last Exam | — | 5.4% |
| MMLU-Pro | — | 81.1% |
| Vectara Hallucination Rate | — | 5.6% |
| GPQA (HELM) | — | 65.9% |

## Multimodal

- Claude Fable 5: 45.6 (#14)
- GPT-4.1: 37.5 (#66)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| LMArena Vision | 1326 | 1210 |
| GeoBench | — | 72% |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
| LMArena Document | 1496 | — |

## Multilingual

- Claude Fable 5: 57.6 (#8)
- GPT-4.1: 49.6 (#128)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| LMArena Non-English | 1483 | 1370 |
| LMArena Chinese | 1546 | 1382 |
| LMArena French | 1509 | 1383 |
| LMArena German | 1486 | 1380 |
| LMArena Japanese | 1507 | 1320 |
| LMArena Korean | 1493 | 1339 |
| LMArena Russian | 1507 | 1377 |
| LMArena Spanish | 1503 | 1375 |

## Instruction Following

- Claude Fable 5: 78.8 (#7)
- GPT-4.1: 71.5 (#148)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| LMArena Instruction Following | 1503 | 1367 |
| IFEval | — | 83.8% |

## Long Context

- Claude Fable 5: 46.3 (#22)
- GPT-4.1: 40.0 (#158)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| LMArena Longer Query | 1510 | 1385 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- Claude Fable 5: 76.1 (#5)
- GPT-4.1: 57.8 (#120)

| Benchmark | Claude Fable 5 | GPT-4.1 |
|---|---|---|
| LMArena Text | 1491 | 1383 |
| LMArena Creative Writing | 1494 | 1363 |
| EQ-Bench Creative Writing | 1943 | 1420 |
| LMArena Multi-Turn | 1505 | 1398 |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1340 | — |

## FAQ

### Is Claude Fable 5 better than GPT-4.1?

Claude Fable 5 is the stronger model overall, scoring 66.3 to 35.7 on the Noometry Index. GPT-4.1 costs 5.7× 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-4.1?

GPT-4.1 is cheaper. It lists at $2 per million input tokens and $8 per million output tokens; Claude Fable 5 lists at $10 and $50.

### Is Claude Fable 5 or GPT-4.1 better for coding?

Claude Fable 5 scores higher on coding benchmarks: 67.0 versus 34.2 in the Noometry coding category.

### Which has the bigger context window?

GPT-4.1 does, with 1.05M tokens against 1M.

### How many benchmarks do Claude Fable 5 and GPT-4.1 share?

34 benchmarks have published results for both models. Claude Fable 5 has 61 scored results on Noometry and GPT-4.1 has 52.