# Claude Fable 5 vs GPT-4o mini

> Claude Fable 5 is the stronger model overall, scoring 66.3 to 25.4 on the Noometry Index. GPT-4o mini costs 76× 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-4o-mini
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Claude Fable 5 scores higher in 10 categories and GPT-4o mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5 leads 88.4 to 9.9.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 100% for Claude Fable 5 and 6.9% for GPT-4o mini.
- GPT-4o mini is cheaper at $0.15 / $0.60 per million input/output tokens, against $10 / $50 for Claude Fable 5.
- Claude Fable 5 accepts more context: 1M tokens versus 128K.

## Snapshot

| | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 66.3 | 25.4 |
| Rank | 5 | 337 |
| Context | 1M | 128K |
| Input $/M | $10 | $0.15 |
| Output $/M | $50 | $0.60 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Fable 5: 67.0 (#4)
- GPT-4o mini: 21.6 (#330)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| WeirdML | 91.9% | 11.8% |
| LMArena Coding | 1519 | 1290 |
| DeepSWE | 69.9% | — |
| FrontierCode | 53.5% | — |
| Aider Polyglot | — | 3.6% |
| LMArena WebDev | 1626 | — |
| FrontierSWE | 47% | — |
| SciCode | 61% | — |
| BigCodeBench Instruct | — | 46.1% |
| LiveBench Coding | — | 43.1% |
| MirrorCode | 63.9% | — |
| BigCodeBench Complete | — | 57.4% |
| ALE-Bench | 2,041 | — |
| HumanEval+ | — | 83.5% |
| MBPP+ | — | 72.2% |

## Agentic & Tool Use

- Claude Fable 5: 54.1 (#2)
- GPT-4o mini: 27.2 (#95)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| APEX-Agents | 63.6% | — |
| Remote Labor Index | 16.1% | — |
| τ²-bench Banking | 39.7% | — |
| PostTrainBench | 41.8% | — |
| BALROG | — | 17.4% |
| GBAEval | 74.5% | — |
| GDP.pdf | 30% | — |
| LMArena Search | 1230 | — |
| Vending-Bench 2 | 5,680 | — |

## Reasoning

- Claude Fable 5: 76.8 (#4)
- GPT-4o mini: 8.7 (#343)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| ARC-AGI-2 | 89.2% | 0% |
| SimpleBench | 81.9% | 10.7% |
| Kagi LLM Benchmark | 91.4% | 28.8% |
| Chess Puzzles | 41% | 0% |
| LMArena Hard Prompts | 1508 | 1267 |
| Mystery Game Puzzles | 52% | 12% |
| DTBench | 98.4% | 54.4% |
| LMCA | 61.1% | 10.4% |
| Epoch Capabilities Index | 163.6 | 126.56 |
| NYT Connections (extended) | 92.7% | — |
| ARC-AGI-1 | 98.5% | — |
| CritPt | 28.6% | — |
| EnigmaEval | 39.3% | — |
| EBR-Bench | 39.5% | — |
| LiveBench Reasoning | — | 32.8% |
| LiveBench Data Analysis | — | 50% |
| Surface Evolver Bench | 95% | — |
| Bench to the Future 3 | 0.13 | — |
| LiveBench | — | 41.3% |
| PIQA | — | 88.7% |

## Math

- Claude Fable 5: 88.4 (#5)
- GPT-4o mini: 9.9 (#307)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 87% | 0.7% |
| OTIS Mock AIME 2024-2025 | 100% | 6.9% |
| LMArena Math | 1519 | 1267 |
| FrontierMath Tier 4 | 90.2% | — |
| ProofBench | 95% | — |
| Omni-MATH | — | 28% |
| LiveBench Math | — | 36.3% |
| MATH Level 5 | — | 52.6% |
| FrontierMath Erdős | 0% | — |
| GSM8K | — | 91.3% |

## Knowledge

- Claude Fable 5: 62.3 (#22)
- GPT-4o mini: 17.7 (#279)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| GPQA Diamond | 85.9% | 37.7% |
| SimpleQA Verified | 70.7% | 8.3% |
| LMArena Expert | 1534 | 1235 |
| MMLU-Pro | — | 60.3% |
| Confabulations | — | 37.2% |
| GPQA (HELM) | — | 36.8% |
| BoolQ | — | 88.7% |
| MMLU | — | 81.8% |

## Multimodal

- Claude Fable 5: 45.6 (#14)
- GPT-4o mini: 25.3 (#117)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| LMArena Vision | 1326 | 1066 |
| Video-MME | — | 64.8% |
| GeoBench | — | 64% |
| VPCT | — | 34% |
| Blueprint-Bench 2 | 38.6% | — |
| Furniture Assembly | 35.8% | — |
| LMArena Document | 1496 | — |

## Multilingual

- Claude Fable 5: 57.6 (#8)
- GPT-4o mini: 42.2 (#194)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| LMArena Non-English | 1483 | 1266 |
| LMArena Chinese | 1546 | 1264 |
| LMArena French | 1509 | 1296 |
| LMArena German | 1486 | 1272 |
| LMArena Japanese | 1507 | 1216 |
| LMArena Korean | 1493 | 1195 |
| LMArena Russian | 1507 | 1275 |
| LMArena Spanish | 1503 | 1276 |

## Instruction Following

- Claude Fable 5: 78.8 (#7)
- GPT-4o mini: 62.1 (#234)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| LMArena Instruction Following | 1503 | 1259 |
| LiveBench Instruction Following | — | 56.8% |
| IFEval | — | 78.2% |

## Long Context

- Claude Fable 5: 46.3 (#22)
- GPT-4o mini: 39.1 (#181)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| LMArena Longer Query | 1510 | 1289 |

## Writing & Preference

- Claude Fable 5: 76.1 (#5)
- GPT-4o mini: 39.7 (#243)

| Benchmark | Claude Fable 5 | GPT-4o mini |
|---|---|---|
| LMArena Text | 1491 | 1286 |
| LMArena Creative Writing | 1494 | 1268 |
| EQ-Bench Creative Writing | 1943 | 873 |
| LMArena Multi-Turn | 1505 | 1285 |
| Short-Story Creative Writing | — | 67.2% |
| WildBench | — | 79.1% |
| EQ-Bench 4 | 1340 | — |
| LiveBench Language | — | 28.6% |

## FAQ

### Is Claude Fable 5 better than GPT-4o mini?

Claude Fable 5 is the stronger model overall, scoring 66.3 to 25.4 on the Noometry Index. GPT-4o mini costs 76× 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-4o mini?

GPT-4o mini is cheaper. It lists at $0.15 per million input tokens and $0.60 per million output tokens; Claude Fable 5 lists at $10 and $50.

### Is Claude Fable 5 or GPT-4o mini better for coding?

Claude Fable 5 scores higher on coding benchmarks: 67.0 versus 21.6 in the Noometry coding category.

### Which has the bigger context window?

Claude Fable 5 does, with 1M tokens against 128K.

### How many benchmarks do Claude Fable 5 and GPT-4o mini share?

32 benchmarks have published results for both models. Claude Fable 5 has 61 scored results on Noometry and GPT-4o mini has 60.