# Claude Sonnet 4.6 vs gpt-oss-20b

> Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 32.5 on the Noometry Index. gpt-oss-20b costs 167× 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-oss-20b
- Last updated: 2026-10-12
- Shared benchmarks: 28

## Summary

- They share 28 benchmarks with published results for both. Claude Sonnet 4.6 scores higher in 9 categories and gpt-oss-20b in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Sonnet 4.6 leads 70.2 to 35.5.
- The biggest single-benchmark swing is Terminal-Bench: 53.4% for Claude Sonnet 4.6 and 3.4% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $3 / $15 for Claude Sonnet 4.6.
- Claude Sonnet 4.6 accepts more context: 1M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 50.3 | 32.5 |
| Rank | 50 | 255 |
| Context | 1M | 131K |
| Input $/M | $3 | $0.018 |
| Output $/M | $15 | $0.09 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 4.6: 46.3 (#67)
- gpt-oss-20b: 37.6 (#192)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| SciCode | 46.8% | 34.4% |
| WeirdML | 66.1% | 40.9% |
| LMArena Coding | 1504 | 1306 |
| ALE-Bench | 1,327 | 566.05 |
| SWE-bench Verified | 75.2% | — |
| DeepSWE | 29.9% | — |
| FrontierCode | 24.3% | — |
| LMArena WebDev | 1522 | — |

## Agentic & Tool Use

- Claude Sonnet 4.6: 39.1 (#28)
- gpt-oss-20b: 9.3 (#154)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | 53.4% | 3.4% |
| APEX-Agents | 43% | — |
| 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-oss-20b: 19.3 (#261)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| CritPt | 3.1% | 1.4% |
| Chess Puzzles | 13% | 4% |
| LMArena Hard Prompts | 1484 | 1274 |
| DTBench | 89.9% | 68% |
| LMCA | 46.5% | 14.5% |
| Epoch Capabilities Index | 152.24 | 137.82 |
| ARC-AGI-2 | 60.4% | — |
| Kagi LLM Benchmark | — | 53.2% |
| NYT Connections (extended) | 80.9% | — |
| ARC-AGI-1 | 86.5% | — |
| Thematic Generalization | 76.3% | — |
| Mystery Game Puzzles | 16% | — |
| ForecastBench | 62 | — |

## Math

- Claude Sonnet 4.6: 52.9 (#49)
- gpt-oss-20b: 39.4 (#103)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 85.8% | 65.3% |
| LMArena Math | 1462 | 1317 |
| ProofBench | 45% | — |
| Omni-MATH | — | 56.5% |
| FrontierMath (Feb 2025 set) | 32.4% | — |
| FrontierMath Tier 4 (v1) | 8.3% | — |

## Knowledge

- Claude Sonnet 4.6: 51.7 (#65)
- gpt-oss-20b: 34.6 (#195)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 87.4% | 60.8% |
| LMArena Expert | 1500 | 1258 |
| SimpleQA Verified | 35.5% | — |
| MMLU-Pro | — | 74% |
| Vectara Hallucination Rate | 10.6% | — |
| GPQA (HELM) | — | 59.4% |

## Multimodal

- Claude Sonnet 4.6: 38.0 (#68)
- gpt-oss-20b: —

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1283 | — |
| Blueprint-Bench 2 | 6.7% | — |
| LMArena Document | 1482 | — |

## Multilingual

- Claude Sonnet 4.6: 54.4 (#41)
- gpt-oss-20b: 42.2 (#197)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1440 | 1268 |
| LMArena Chinese | 1491 | 1314 |
| LMArena German | 1428 | 1255 |
| LMArena Japanese | 1420 | 1244 |
| LMArena Korean | 1411 | 1236 |
| LMArena Russian | 1440 | 1278 |
| LMArena Spanish | 1464 | 1267 |
| LMArena French | 1465 | — |

## Instruction Following

- Claude Sonnet 4.6: 77.4 (#25)
- gpt-oss-20b: 61.8 (#240)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1475 | 1236 |
| IFEval | — | 73.2% |

## Long Context

- Claude Sonnet 4.6: 45.3 (#44)
- gpt-oss-20b: 37.9 (#209)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1479 | 1250 |

## Writing & Preference

- Claude Sonnet 4.6: 70.2 (#22)
- gpt-oss-20b: 35.5 (#265)

| Benchmark | Claude Sonnet 4.6 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1458 | 1287 |
| LMArena Creative Writing | 1435 | 1201 |
| EQ-Bench Creative Writing | 1810 | 666 |
| LMArena Multi-Turn | 1464 | 1268 |
| WildBench | — | 73.7% |
| EQ-Bench 4 | 1207 | — |

## FAQ

### Is Claude Sonnet 4.6 better than gpt-oss-20b?

Claude Sonnet 4.6 is the stronger model overall, scoring 50.3 to 32.5 on the Noometry Index. gpt-oss-20b costs 167× 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-oss-20b?

gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Claude Sonnet 4.6 lists at $3 and $15.

### Is Claude Sonnet 4.6 or gpt-oss-20b better for coding?

Claude Sonnet 4.6 scores higher on coding benchmarks: 46.3 versus 37.6 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 4.6 does, with 1M tokens against 131K.

### How many benchmarks do Claude Sonnet 4.6 and gpt-oss-20b share?

28 benchmarks have published results for both models. Claude Sonnet 4.6 has 57 scored results on Noometry and gpt-oss-20b has 34.