Model comparison
Claude 3.5 Sonnet vs gpt-oss-20b
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 32.5 on the Noometry Index.
Last verified . 27 shared benchmarks.
Summary
- They share 27 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 7 categories and gpt-oss-20b in 2 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude 3.5 Sonnet leads 32.3 to 9.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 65.3% for gpt-oss-20b.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| Claude 3.5 Sonnet | gpt-oss-20b | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.6 | 32.5 |
| Released | 2024-06-20 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | — | 131K |
| Max output | — | 16K |
| Input $ / M tokens | — | $0.018 |
| Output $ / M tokens | — | $0.09 |
| Results tracked | 60 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.0 (#165), gpt-oss-20b: 37.6 (#192)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| WeirdML | 40% | 40.9% |
| LMArena Coding | 1342 | 1306 |
| Aider Polyglot | 51.6% | — |
| SciCode | — | 34.4% |
| GSO | 4.6% | — |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
| ALE-Bench | — | 566.05 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 32.3 (#67), gpt-oss-20b: 9.3 (#154)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| BALROG | 32.6% | — |
| METR Time Horizons | 45.2% | — |
Reasoning Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 23.1 (#183), gpt-oss-20b: 19.3 (#261)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1305 | 1274 |
| DTBench | 67.8% | 68% |
| Epoch Capabilities Index | 133.55 | 137.82 |
| SimpleBench | 41.4% | — |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| EnigmaEval | 0.9% | — |
| LiveBench Reasoning | 56.7% | — |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 14.5% |
| ForecastBench | 60.7 | — |
| LiveBench | 59% | — |
Math gpt-oss-20b leads
Claude 3.5 Sonnet: 19.2 (#288), gpt-oss-20b: 39.4 (#103)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 65.3% |
| Omni-MATH | 27.6% | 56.5% |
| LMArena Math | 1307 | 1317 |
| LiveBench Math | 52.3% | — |
| MATH Level 5 | 56.9% | — |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge gpt-oss-20b leads
Claude 3.5 Sonnet: 28.6 (#245), gpt-oss-20b: 34.6 (#195)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 55.3% | 60.8% |
| MMLU-Pro | 77.7% | 74% |
| GPQA (HELM) | 56.5% | 59.4% |
| LMArena Expert | 1265 | 1258 |
| Humanity's Last Exam | 4.1% | — |
| Confabulations | 19.9% | — |
| MMLU | 87.3% | — |
Multimodal Not comparable
Claude 3.5 Sonnet: 26.5 (#120), gpt-oss-20b: —
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1125 | — |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
Multilingual Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 43.2 (#185), gpt-oss-20b: 42.2 (#197)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1283 | 1268 |
| LMArena Chinese | 1272 | 1314 |
| LMArena German | 1297 | 1255 |
| LMArena Japanese | 1234 | 1244 |
| LMArena Korean | 1200 | 1236 |
| LMArena Russian | 1306 | 1278 |
| LMArena Spanish | 1290 | 1267 |
| LMArena French | 1305 | — |
Instruction Following Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 68.8 (#182), gpt-oss-20b: 61.8 (#240)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| IFEval | 85.5% | 73.2% |
| LMArena Instruction Following | 1297 | 1236 |
| LiveBench Instruction Following | 69.3% | — |
Long Context Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 39.9 (#167), gpt-oss-20b: 37.9 (#209)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1311 | 1250 |
Writing & Preference Claude 3.5 Sonnet leads
Claude 3.5 Sonnet: 52.9 (#164), gpt-oss-20b: 35.5 (#265)
| Benchmark | Claude 3.5 Sonnet | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1298 | 1287 |
| LMArena Creative Writing | 1292 | 1201 |
| EQ-Bench Creative Writing | 1451 | 666 |
| WildBench | 79.2% | 73.7% |
| LMArena Multi-Turn | 1326 | 1268 |
| Short-Story Creative Writing | 80.3% | — |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than gpt-oss-20b?
Claude 3.5 Sonnet is the stronger model overall, scoring 34.6 to 32.5 on the Noometry Index.
Is Claude 3.5 Sonnet or gpt-oss-20b better for coding?
Claude 3.5 Sonnet scores higher on coding benchmarks: 39.0 versus 37.6 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and gpt-oss-20b share?
27 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and gpt-oss-20b has 34.