# Claude Sonnet 4.6 vs o4-mini

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 41.6 on the Noometry Index. o4-mini costs 3.1× 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-o4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 35

## Summary

- They share 35 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 8 categories and o4-mini in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 4.6 leads 46.1 to 24.6.
- The biggest single-benchmark swing is ARC-AGI-2: 60.4% for Claude Sonnet 4.6 and 6.1% for o4-mini.
- o4-mini is cheaper at $1.10 / $4.40 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 200K.

## Snapshot

| | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 41.6 |
| Rank | 50 | 132 |
| Context | 1M | 200K |
| Input $/M | $3 | $1.10 |
| Output $/M | $15 | $4.40 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Sonnet 4.6: 46.3 (#67)
- o4-mini: 40.9 (#127)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| WeirdML | 66.1% | 52.6% |
| LMArena Coding | 1504 | 1368 |
| ALE-Bench | 1,327 | 826.17 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| SWE-bench Verified (bash only) | — | 45% |
| Aider Polyglot | — | 72% |
| LMArena WebDev | 1522 | — |
| SciCode | 46.8% | — |
| GSO | — | 3.6% |
| CadEval | — | 62% |
| AlgoTune | — | 1.72 |

## Agentic & Tool Use

- Claude Sonnet 4.6: 39.1 (#28)
- o4-mini: 32.6 (#61)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| Terminal-Bench | 53.4% | — |
| APEX-Agents | 43% | — |
| Berkeley Function Calling Leaderboard | — | 53.2% |
| OSWorld 2.0 | 9.3% | — |
| GDPval | — | 25.3% |
| DeepResearch Bench | 54.9% | — |
| OSWorld | 72.1% | — |
| ExploitBench | 23.6% | — |
| GBAEval | 48.8% | — |
| GDP.pdf | 18% | — |
| LMArena Search | 1221 | — |
| METR Time Horizons | — | 63.9% |
| Vending-Bench 2 | 7,204 | — |

## Reasoning

- Claude Sonnet 4.6: 46.1 (#45)
- o4-mini: 24.6 (#162)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| ARC-AGI-2 | 60.4% | 6.1% |
| ARC-AGI-1 | 86.5% | 58.7% |
| CritPt | 3.1% | 0.6% |
| Chess Puzzles | 13% | 26% |
| LMArena Hard Prompts | 1484 | 1351 |
| Mystery Game Puzzles | 16% | 5% |
| DTBench | 89.9% | 77.6% |
| LMCA | 46.5% | 26.5% |
| Epoch Capabilities Index | 152.24 | 145.64 |
| ForecastBench | 62 | 61.8 |
| SimpleBench | — | 38.7% |
| Kagi LLM Benchmark | — | 67.6% |
| NYT Connections (extended) | 80.9% | — |
| EnigmaEval | — | 9.2% |
| Thematic Generalization | 76.3% | — |

## Math

- Claude Sonnet 4.6: 52.9 (#49)
- o4-mini: 40.8 (#89)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 81.7% |
| LMArena Math | 1462 | 1389 |
| FrontierMath (Feb 2025 set) | 32.4% | 24.8% |
| FrontierMath Tier 4 (v1) | 8.3% | 6.3% |
| FrontierMath (Tiers 1-3) | — | 36.1% |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | 45% | — |
| Omni-MATH | — | 72% |
| MATH Level 5 | — | 97.8% |

## Knowledge

- Claude Sonnet 4.6: 51.7 (#65)
- o4-mini: 43.6 (#91)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| GPQA Diamond | 87.4% | 79.6% |
| SimpleQA Verified | 35.5% | 19.6% |
| Vectara Hallucination Rate | 10.6% | 18.6% |
| LMArena Expert | 1500 | 1343 |
| Humanity's Last Exam | — | 18.1% |
| MMLU-Pro | — | 82% |
| Confabulations | — | 15.8% |
| GPQA (HELM) | — | 73.5% |

## Multimodal

- Claude Sonnet 4.6: 38.0 (#68)
- o4-mini: 40.2 (#49)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| LMArena Vision | 1283 | 1194 |
| GeoBench | — | 64% |
| VPCT | — | 57.5% |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |

## Multilingual

- Claude Sonnet 4.6: 54.4 (#41)
- o4-mini: 47.0 (#154)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| LMArena Non-English | 1440 | 1337 |
| LMArena Chinese | 1491 | 1354 |
| LMArena French | 1465 | 1364 |
| LMArena German | 1428 | 1336 |
| LMArena Japanese | 1420 | 1308 |
| LMArena Korean | 1411 | 1312 |
| LMArena Russian | 1440 | 1334 |
| LMArena Spanish | 1464 | 1347 |

## Instruction Following

- Claude Sonnet 4.6: 77.4 (#25)
- o4-mini: 75.2 (#68)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| LMArena Instruction Following | 1475 | 1321 |
| IFEval | — | 92.8% |

## Long Context

- Claude Sonnet 4.6: 45.3 (#44)
- o4-mini: 45.5 (#33)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| LMArena Longer Query | 1479 | 1315 |
| Fiction.LiveBench | — | 77.8% |

## Writing & Preference

- Claude Sonnet 4.6: 70.2 (#22)
- o4-mini: 54.0 (#152)

| Benchmark | Claude Sonnet 4.6 | o4-mini |
|---|---|---|
| LMArena Text | 1458 | 1353 |
| LMArena Creative Writing | 1435 | 1294 |
| LMArena Multi-Turn | 1464 | 1350 |
| Short-Story Creative Writing | — | 75% |
| EQ-Bench Creative Writing | 1810 | — |
| WildBench | — | 85.4% |
| EQ-Bench 4 | 1207 | — |

## FAQ

### Is Claude Sonnet 4.6 better than o4-mini?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 41.6 on the Noometry Index. o4-mini costs 3.1× 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 o4-mini?

o4-mini is cheaper. It lists at $1.10 per million input tokens and $4.40 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Claude Sonnet 4.6 or o4-mini better for coding?

Claude Sonnet 4.6 scores higher on coding benchmarks: 46.3 versus 40.9 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 4.6 does, with 1M tokens against 200K.

### How many benchmarks do Claude Sonnet 4.6 and o4-mini share?

35 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and o4-mini has 60.