# Claude Sonnet 4.6 vs GPT-4.1 mini

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 33.6 on the Noometry Index. GPT-4.1 mini costs 8.6× 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-4-1-mini
- Last updated: 2026-10-12
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 10 categories and GPT-4.1 mini in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 10.8.
- The biggest single-benchmark swing is ARC-AGI-1: 86.5% for Claude Sonnet 4.6 and 3.5% for GPT-4.1 mini.
- GPT-4.1 mini is cheaper at $0.40 / $1.60 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 1M.

## Snapshot

| | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 33.6 |
| Rank | 50 | 240 |
| Context | 1M | 1.05M |
| Input $/M | $3 | $0.40 |
| Output $/M | $15 | $1.60 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 4.6: 46.3 (#67)
- GPT-4.1 mini: 30.6 (#293)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| SciCode | 46.8% | 40.4% |
| WeirdML | 66.1% | 37.6% |
| LMArena Coding | 1504 | 1367 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 23.9% |
| Aider Polyglot | — | 32.4% |
| LMArena WebDev | 1522 | — |
| BigCodeBench Instruct | — | 48.9% |
| CadEval | — | 16% |
| ALE-Bench | 1,327 | — |

## Agentic & Tool Use

- Claude Sonnet 4.6: 39.1 (#28)
- GPT-4.1 mini: 33.3 (#55)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| Berkeley Function Calling Leaderboard | — | 50.5% |
| OSWorld 2.0 | 9.3% | — |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| Vending-Bench 2 | 7,204 | — |

## Reasoning

- Claude Sonnet 4.6: 46.1 (#45)
- GPT-4.1 mini: 10.8 (#340)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| ARC-AGI-2 | 60.4% | 0% |
| ARC-AGI-1 | 86.5% | 3.5% |
| CritPt | 3.1% | 0% |
| Chess Puzzles | 13% | 7% |
| LMArena Hard Prompts | 1484 | 1349 |
| Mystery Game Puzzles | 16% | 7% |
| DTBench | 89.9% | 68.8% |
| LMCA | 46.5% | 21.1% |
| Epoch Capabilities Index | 152.24 | 135.01 |
| Kagi LLM Benchmark | — | 48.6% |
| NYT Connections (extended) | 80.9% | — |
| Thematic Generalization | 76.3% | — |
| ForecastBench | 62 | — |

## Math

- Claude Sonnet 4.6: 52.9 (#49)
- GPT-4.1 mini: 24.1 (#270)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 44.7% |
| LMArena Math | 1462 | 1343 |
| FrontierMath (Feb 2025 set) | 32.4% | 4.5% |
| FrontierMath (Tiers 1-3) | — | 6.7% |
| ProofBench | 45% | — |
| Omni-MATH | — | 49.1% |
| MATH Level 5 | — | 87.3% |
| FrontierMath Tier 4 (v1) | 8.3% | — |

## Knowledge

- Claude Sonnet 4.6: 51.7 (#65)
- GPT-4.1 mini: 34.7 (#194)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| GPQA Diamond | 87.4% | 65.8% |
| SimpleQA Verified | 35.5% | 12.7% |
| LMArena Expert | 1500 | 1338 |
| MMLU-Pro | — | 78.3% |
| Vectara Hallucination Rate | 10.6% | — |
| GPQA (HELM) | — | 61.4% |

## Multimodal

- Claude Sonnet 4.6: 38.0 (#68)
- GPT-4.1 mini: 35.8 (#82)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| LMArena Vision | 1283 | 1181 |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |

## Multilingual

- Claude Sonnet 4.6: 54.4 (#41)
- GPT-4.1 mini: 45.7 (#166)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| LMArena Non-English | 1440 | 1318 |
| LMArena Chinese | 1491 | 1329 |
| LMArena French | 1465 | 1358 |
| LMArena German | 1428 | 1351 |
| LMArena Japanese | 1420 | 1290 |
| LMArena Korean | 1411 | 1298 |
| LMArena Russian | 1440 | 1324 |
| LMArena Spanish | 1464 | 1319 |

## Instruction Following

- Claude Sonnet 4.6: 77.4 (#25)
- GPT-4.1 mini: 73.7 (#118)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| LMArena Instruction Following | 1475 | 1333 |
| IFEval | — | 90.4% |

## Long Context

- Claude Sonnet 4.6: 45.3 (#44)
- GPT-4.1 mini: 31.8 (#275)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| LMArena Longer Query | 1479 | 1344 |
| Fiction.LiveBench | — | 44.4% |

## Writing & Preference

- Claude Sonnet 4.6: 70.2 (#22)
- GPT-4.1 mini: 48.6 (#199)

| Benchmark | Claude Sonnet 4.6 | GPT-4.1 mini |
|---|---|---|
| LMArena Text | 1458 | 1340 |
| LMArena Creative Writing | 1435 | 1300 |
| EQ-Bench Creative Writing | 1810 | 1147 |
| LMArena Multi-Turn | 1464 | 1354 |
| WildBench | — | 83.8% |
| EQ-Bench 4 | 1207 | — |

## FAQ

### Is Claude Sonnet 4.6 better than GPT-4.1 mini?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 33.6 on the Noometry Index. GPT-4.1 mini costs 8.6× 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-4.1 mini?

GPT-4.1 mini is cheaper. It lists at $0.40 per million input tokens and $1.60 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Claude Sonnet 4.6 or GPT-4.1 mini better for coding?

Claude Sonnet 4.6 scores higher on coding benchmarks: 46.3 versus 30.6 in the Noometry coding category.

### Which has the bigger context window?

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

### How many benchmarks do Claude Sonnet 4.6 and GPT-4.1 mini share?

33 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and GPT-4.1 mini has 47.