# Claude Sonnet 4.6 vs GPT-5.1

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 49.0 on the Noometry Index. GPT-5.1 costs 1.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-sonnet-4-6-vs-gpt-5-1
- Last updated: 2026-10-10
- Shared benchmarks: 43

## Summary

- They share 43 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 6 categories and GPT-5.1 in 4 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where GPT-5.1 leads 44.8 to 38.0.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Claude Sonnet 4.6 and 17.6% for GPT-5.1.
- GPT-5.1 is cheaper at $1.25 / $10 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 400K.

## Snapshot

| | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 49.0 |
| Rank | 50 | 53 |
| Context | 1M | 400K |
| Input $/M | $3 | $1.25 |
| Output $/M | $15 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 4.6: 46.3 (#67)
- GPT-5.1: 46.4 (#66)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| SWE-bench Verified | 75.2% | 68% |
| LMArena WebDev | 1522 | 1395 |
| SciCode | 46.8% | 43.3% |
| WeirdML | 66.1% | 60.8% |
| LMArena Coding | 1504 | 1454 |
| ALE-Bench | 1,327 | 1,192 |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 66% |
| GSO | — | 13.7% |
| LiveBench Coding | — | 72.5% |

## Agentic & Tool Use

- Claude Sonnet 4.6: 39.1 (#28)
- GPT-5.1: 32.7 (#60)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| Terminal-Bench | 53.4% | 47.6% |
| DeepResearch Bench | 54.9% | 42.8% |
| LMArena Search | 1221 | 1199 |
| Vending-Bench 2 | 7,204 | 1,473 |
| APEX-Agents | 43% | — |
| OSWorld 2.0 | 9.3% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |

## Reasoning

- Claude Sonnet 4.6: 46.1 (#45)
- GPT-5.1: 39.8 (#58)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 60.4% | 17.6% |
| ARC-AGI-1 | 86.5% | 72.8% |
| CritPt | 3.1% | 4.9% |
| Chess Puzzles | 13% | 32% |
| LMArena Hard Prompts | 1484 | 1457 |
| Mystery Game Puzzles | 16% | 19% |
| DTBench | 89.9% | 90.1% |
| LMCA | 46.5% | 43.9% |
| Epoch Capabilities Index | 152.24 | 149.64 |
| ForecastBench | 62 | 58.1 |
| SimpleBench | — | 53.2% |
| NYT Connections (extended) | 80.9% | — |
| EnigmaEval | — | 11.2% |
| Thematic Generalization | 76.3% | — |
| LiveBench Reasoning | — | 95.8% |
| LiveBench Data Analysis | — | 72.1% |
| LiveBench | — | 78.8% |

## Math

- Claude Sonnet 4.6: 52.9 (#49)
- GPT-5.1: 52.2 (#51)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 88.6% |
| LMArena Math | 1462 | 1447 |
| FrontierMath (Feb 2025 set) | 32.4% | 31% |
| FrontierMath Tier 4 (v1) | 8.3% | 12.5% |
| ProofBench | 45% | — |
| Omni-MATH | — | 46.4% |
| LiveBench Math | — | 94.5% |

## Knowledge

- Claude Sonnet 4.6: 51.7 (#65)
- GPT-5.1: 50.6 (#71)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 87.4% | 87.6% |
| SimpleQA Verified | 35.5% | 48% |
| Vectara Hallucination Rate | 10.6% | 10.9% |
| LMArena Expert | 1500 | 1470 |
| Humanity's Last Exam | — | 23.7% |
| MMLU-Pro | — | 57.9% |
| GPQA (HELM) | — | 44.2% |

## Multimodal

- Claude Sonnet 4.6: 38.0 (#68)
- GPT-5.1: 44.8 (#19)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1283 | 1250 |
| LMArena Document | 1482 | 1403 |
| VPCT | — | 58.7% |
| Blueprint-Bench 2 | 6.7% | — |

## Multilingual

- Claude Sonnet 4.6: 54.4 (#41)
- GPT-5.1: 53.8 (#56)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1440 | 1431 |
| LMArena Chinese | 1491 | 1495 |
| LMArena French | 1465 | 1450 |
| LMArena German | 1428 | 1438 |
| LMArena Japanese | 1420 | 1453 |
| LMArena Korean | 1411 | 1401 |
| LMArena Russian | 1440 | 1435 |
| LMArena Spanish | 1464 | 1433 |

## Instruction Following

- Claude Sonnet 4.6: 77.4 (#25)
- GPT-5.1: 83.9 (#1)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| LMArena Instruction Following | 1475 | 1443 |
| LiveBench Instruction Following | — | 93.3% |
| IFEval | — | 93.5% |

## Long Context

- Claude Sonnet 4.6: 45.3 (#44)
- GPT-5.1: 47.6 (#14)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1479 | 1447 |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |

## Writing & Preference

- Claude Sonnet 4.6: 70.2 (#22)
- GPT-5.1: 64.5 (#55)

| Benchmark | Claude Sonnet 4.6 | GPT-5.1 |
|---|---|---|
| LMArena Text | 1458 | 1443 |
| LMArena Creative Writing | 1435 | 1427 |
| LMArena Multi-Turn | 1464 | 1450 |
| EQ-Bench Creative Writing | 1810 | — |
| WildBench | — | 86.3% |
| EQ-Bench 4 | 1207 | — |
| LiveBench Language | — | 80.2% |

## FAQ

### Is Claude Sonnet 4.6 better than GPT-5.1?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 49.0 on the Noometry Index. GPT-5.1 costs 1.7× less per token, which makes it the better buy when Claude Sonnet 4.6's lead doesn't matter for your workload.

### Which is cheaper, Claude Sonnet 4.6 or GPT-5.1?

GPT-5.1 is cheaper. It lists at $1.25 per million input tokens and $10 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Claude Sonnet 4.6 or GPT-5.1 better for coding?

They score almost the same on coding (46.3 vs 46.4); test both on your own repository before choosing.

### Which has the bigger context window?

Claude Sonnet 4.6 does, with 1M tokens against 400K.

### How many benchmarks do Claude Sonnet 4.6 and GPT-5.1 share?

43 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GPT-5.1 has 63.