# Claude Sonnet 5 vs GPT-5.4 mini

> Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 45.0 on the Noometry Index. GPT-5.4 mini costs 2.4× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-sonnet-5-vs-gpt-5-4-mini
- Last updated: 2026-10-12
- Shared benchmarks: 38

## Summary

- They share 38 benchmarks with published results for both. Claude Sonnet 5 scores higher in 10 categories and GPT-5.4 mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5 leads 66.2 to 45.5.
- The biggest single-benchmark swing is ProofBench: 77% for Claude Sonnet 5 and 21% for GPT-5.4 mini.
- GPT-5.4 mini is cheaper at $0.75 / $4.50 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 400K.

## Snapshot

| | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 54.6 | 45.0 |
| Rank | 29 | 76 |
| Context | 1M | 400K |
| Input $/M | $2 | $0.75 |
| Output $/M | $10 | $4.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 5: 55.5 (#26)
- GPT-5.4 mini: 45.2 (#72)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| FrontierCode | 42.7% | 27% |
| LMArena WebDev | 1541 | 1397 |
| SciCode | 54.3% | 49.9% |
| WeirdML | 68.8% | 60.3% |
| LMArena Coding | 1483 | 1438 |
| ALE-Bench | 1,463 | 1,189 |
| DeepSWE | 53.8% | — |
| CursorBench | 34.1% | — |
| GSO | 37.3% | — |

## Agentic & Tool Use

- Claude Sonnet 5: 42.8 (#18)
- GPT-5.4 mini: 29.9 (#81)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| APEX-Agents | 54.5% | — |
| DeepResearch Bench | — | 36.3% |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
| Vending-Bench 2 | 6,378 | — |

## Reasoning

- Claude Sonnet 5: 49.1 (#39)
- GPT-5.4 mini: 30.4 (#85)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| NYT Connections (extended) | 75.1% | 61.8% |
| CritPt | 16.9% | 10% |
| Chess Puzzles | 35% | 24% |
| LMArena Hard Prompts | 1461 | 1424 |
| Mystery Game Puzzles | 35% | 11% |
| DTBench | 92.5% | 80% |
| LMCA | 50% | 40.8% |
| Epoch Capabilities Index | 156.21 | 148.84 |
| ForecastBench | 61.1 | 57 |
| ARC-AGI-2 | — | 18.9% |
| SimpleBench | 60.6% | — |
| Kagi LLM Benchmark | — | 37.9% |
| ARC-AGI-1 | — | 63.7% |
| Thematic Generalization | — | 61.7% |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |

## Math

- Claude Sonnet 5: 66.2 (#27)
- GPT-5.4 mini: 45.5 (#75)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| FrontierMath (Tiers 1-3) | 65.6% | 51.2% |
| FrontierMath Tier 4 | 29.3% | 9.8% |
| OTIS Mock AIME 2024-2025 | 94.7% | 88.9% |
| ProofBench | 77% | 21% |
| LMArena Math | 1467 | 1419 |
| FrontierMath (Feb 2025 set) | — | 28.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Claude Sonnet 5: 55.6 (#47)
- GPT-5.4 mini: 51.5 (#67)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| GPQA Diamond | 90.5% | 86.9% |
| SimpleQA Verified | 33.7% | 29.4% |
| LMArena Expert | 1490 | 1435 |
| Vectara Hallucination Rate | — | 5.5% |

## Multimodal

- Claude Sonnet 5: 42.4 (#31)
- GPT-5.4 mini: 39.7 (#56)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Vision | 1274 | 1245 |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |

## Multilingual

- Claude Sonnet 5: 53.8 (#55)
- GPT-5.4 mini: 51.9 (#96)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Non-English | 1431 | 1405 |
| LMArena Chinese | 1477 | 1446 |
| LMArena French | 1460 | 1440 |
| LMArena German | 1440 | 1409 |
| LMArena Japanese | 1422 | 1374 |
| LMArena Korean | 1411 | 1368 |
| LMArena Russian | 1451 | 1417 |
| LMArena Spanish | 1437 | 1405 |

## Instruction Following

- Claude Sonnet 5: 76.3 (#41)
- GPT-5.4 mini: 74.1 (#102)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Instruction Following | 1452 | 1405 |

## Long Context

- Claude Sonnet 5: 44.8 (#55)
- GPT-5.4 mini: 43.0 (#112)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Longer Query | 1463 | 1407 |

## Writing & Preference

- Claude Sonnet 5: 69.2 (#25)
- GPT-5.4 mini: 64.0 (#58)

| Benchmark | Claude Sonnet 5 | GPT-5.4 mini |
|---|---|---|
| LMArena Text | 1442 | 1412 |
| LMArena Creative Writing | 1416 | 1370 |
| EQ-Bench Creative Writing | 1794 | 1665 |
| LMArena Multi-Turn | 1454 | 1429 |
| EQ-Bench 4 | 1236 | — |

## FAQ

### Is Claude Sonnet 5 better than GPT-5.4 mini?

Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 45.0 on the Noometry Index. GPT-5.4 mini costs 2.4× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

### Which is cheaper, Claude Sonnet 5 or GPT-5.4 mini?

GPT-5.4 mini is cheaper. It lists at $0.75 per million input tokens and $4.50 per million output tokens; Claude Sonnet 5 lists at $2 and $10.

### Is Claude Sonnet 5 or GPT-5.4 mini better for coding?

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 45.2 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 5 does, with 1M tokens against 400K.

### How many benchmarks do Claude Sonnet 5 and GPT-5.4 mini share?

38 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and GPT-5.4 mini has 46.