Model comparison
Claude 3.7 Sonnet vs gpt-oss-20b
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 32.1 on the Noometry Index.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude 3.7 Sonnet scores higher in 7 categories and gpt-oss-20b in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude 3.7 Sonnet leads 33.9 to 9.2.
- The biggest single-benchmark swing is Omni-MATH: 33% for Claude 3.7 Sonnet and 56.5% for gpt-oss-20b.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| Claude 3.7 Sonnet | gpt-oss-20b | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 39.4 | 32.1 |
| Released | 2025-02-24 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 58 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 41.1 (#120), gpt-oss-20b: 37.1 (#185)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Coding | 1361 | 1307 |
| SWE-bench Verified | 61% | — |
| SWE-bench Verified (bash only) | 52.8% | — |
| Aider Polyglot | 64.9% | — |
| SciCode | — | 34.4% |
| GSO | 3.8% | — |
| WeirdML | — | 40.9% |
| LiveBench Coding | 74.5% | — |
| CadEval | 54% | — |
| ALE-Bench | — | 566.05 |
Agentic & Tool Use Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 33.9 (#46), gpt-oss-20b: 9.2 (#146)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| TheAgentCompany | 30.9% | — |
| Cybench | 20% | — |
| DeepResearch Bench | 43.6% | — |
| OSWorld | 35.8% | — |
| METR Time Horizons | 60% | — |
Reasoning Too close to call
Claude 3.7 Sonnet: 18.7 (#252), gpt-oss-20b: 18.8 (#251)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1333 | 1274 |
| Epoch Capabilities Index | 141.16 | 137.81 |
| ARC-AGI-2 | 0.9% | — |
| SimpleBench | 46.4% | — |
| Kagi LLM Benchmark | — | 53.2% |
| ARC-AGI-1 | 28.6% | — |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| EnigmaEval | 4.2% | — |
| LiveBench Reasoning | 87.8% | — |
| DTBench | — | 68% |
| LiveBench Data Analysis | 74% | — |
| LMCA | — | 14.5% |
| ForecastBench | 61.8 | — |
| LiveBench | 76.1% | — |
Math gpt-oss-20b leads
Claude 3.7 Sonnet: 35.7 (#137), gpt-oss-20b: 37.6 (#93)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 57.8% | 65.3% |
| Omni-MATH | 33% | 56.5% |
| LMArena Math | 1336 | 1317 |
| LiveBench Math | 79% | — |
| MATH Level 5 | 91.2% | — |
| FrontierMath (Feb 2025 set) | 4.1% | — |
Knowledge Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 39.7 (#117), gpt-oss-20b: 34.6 (#186)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 79.7% | 60.8% |
| MMLU-Pro | 78.4% | 74% |
| GPQA (HELM) | 60.8% | 59.4% |
| LMArena Expert | 1321 | 1258 |
| Humanity's Last Exam | 8% | — |
| Confabulations | 14.7% | — |
Multimodal Not comparable
Claude 3.7 Sonnet: 33.1 (#90), gpt-oss-20b: —
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1169 | — |
| GeoBench | 68% | — |
| VPCT | 39% | — |
| SpatialViz-Bench | 33.9% | — |
Multilingual Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 44.3 (#174), gpt-oss-20b: 42.4 (#192)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1296 | 1269 |
| LMArena Chinese | 1299 | 1311 |
| LMArena German | 1301 | 1249 |
| LMArena Japanese | 1268 | 1245 |
| LMArena Korean | 1249 | 1239 |
| LMArena Russian | 1311 | 1279 |
| LMArena Spanish | 1299 | 1267 |
| LMArena French | 1302 | — |
Instruction Following Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 73.1 (#120), gpt-oss-20b: 62.0 (#235)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| IFEval | 83.4% | 73.2% |
| LMArena Instruction Following | 1352 | 1236 |
| LiveBench Instruction Following | 81.3% | — |
Long Context Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 50.2 (#10), gpt-oss-20b: 37.9 (#203)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1373 | 1251 |
| Fiction.LiveBench | 83.3% | — |
Writing & Preference Claude 3.7 Sonnet leads
Claude 3.7 Sonnet: 54.6 (#145), gpt-oss-20b: 35.8 (#260)
| Benchmark | Claude 3.7 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1314 | 1288 |
| LMArena Creative Writing | 1332 | 1201 |
| EQ-Bench Creative Writing | 1412 | 666 |
| WildBench | 81.4% | 73.7% |
| LMArena Multi-Turn | 1339 | 1268 |
| Short-Story Creative Writing | 81.1% | — |
| LiveBench Language | 59.9% | — |
Frequently asked questions
Is Claude 3.7 Sonnet better than gpt-oss-20b?
Claude 3.7 Sonnet is the stronger model overall, scoring 39.4 to 32.1 on the Noometry Index.
Is Claude 3.7 Sonnet or gpt-oss-20b better for coding?
Claude 3.7 Sonnet scores higher on coding benchmarks: 41.1 versus 37.1 in the Noometry coding category.
How many benchmarks do Claude 3.7 Sonnet and gpt-oss-20b share?
25 benchmarks have published results for both models. Claude 3.7 Sonnet has 58 scored results on Noometry and gpt-oss-20b has 34.