Model comparison
Claude Opus 4.1 vs gpt-oss-20b
Claude Opus 4.1 is the stronger model overall, scoring 40.8 to 32.1 on the Noometry Index. gpt-oss-20b costs 833× less per token, which makes it the better buy when Claude Opus 4.1's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Summary
- They share 25 benchmarks with published results for both. Claude Opus 4.1 scores higher in 8 categories and gpt-oss-20b in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Claude Opus 4.1 leads 62.6 to 35.8.
- The biggest single-benchmark swing is Terminal-Bench: 38% for Claude Opus 4.1 and 3.4% for gpt-oss-20b.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $15 / $75 for Claude Opus 4.1.
- Claude Opus 4.1 accepts more context: 200K tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| Claude Opus 4.1 | gpt-oss-20b | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 40.8 | 32.1 |
| Released | 2025-08-05 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 32K | 16K |
| Input $ / M tokens | $15 | $0.018 |
| Output $ / M tokens | $75 | $0.09 |
| Results tracked | 48 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4.1 leads
Claude Opus 4.1: 44.3 (#69), gpt-oss-20b: 37.1 (#185)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| WeirdML | 45.9% | 40.9% |
| LMArena Coding | 1479 | 1307 |
| ALE-Bench | 674.77 | 566.05 |
| SWE-bench Verified | 73.3% | — |
| LMArena WebDev | 1390 | — |
| SciCode | — | 34.4% |
| AlgoTune | 1.34 | — |
Agentic & Tool Use Claude Opus 4.1 leads
Claude Opus 4.1: 34.8 (#38), gpt-oss-20b: 9.2 (#146)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | 38% | 3.4% |
| GDPval | 43.6% | — |
| Cybench | 42% | — |
| DeepResearch Bench | 48.3% | — |
| LMArena Search | 1148 | — |
| METR Time Horizons | 66.8% | — |
Reasoning Claude Opus 4.1 leads
Claude Opus 4.1: 31.6 (#70), gpt-oss-20b: 18.8 (#251)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| Chess Puzzles | 7% | 4% |
| LMArena Hard Prompts | 1443 | 1274 |
| DTBench | 80% | 68% |
| LMCA | 37.1% | 14.5% |
| Epoch Capabilities Index | 144.11 | 137.81 |
| SimpleBench | 60% | — |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| EnigmaEval | 7.2% | — |
| EBR-Bench | 7.9% | — |
| Mystery Game Puzzles | 21% | — |
| ForecastBench | 62 | — |
Math gpt-oss-20b leads
Claude Opus 4.1: 22.0 (#267), gpt-oss-20b: 37.6 (#93)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 68.9% | 65.3% |
| LMArena Math | 1430 | 1317 |
| FrontierMath (Tiers 1-3) | 12.6% | — |
| FrontierMath Tier 4 | 2.4% | — |
| Omni-MATH | — | 56.5% |
| FrontierMath (Feb 2025 set) | 7.2% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Opus 4.1 leads
Claude Opus 4.1: 41.9 (#93), gpt-oss-20b: 34.6 (#186)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 77.3% | 60.8% |
| LMArena Expert | 1439 | 1258 |
| Humanity's Last Exam | 11.5% | — |
| MMLU-Pro | — | 74% |
| Confabulations | 17.1% | — |
| Vectara Hallucination Rate | 11.8% | — |
| GPQA (HELM) | — | 59.4% |
Multimodal Not comparable
Claude Opus 4.1: 26.2 (#114), gpt-oss-20b: —
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| VPCT | 35% | — |
Multilingual Claude Opus 4.1 leads
Claude Opus 4.1: 52.1 (#90), gpt-oss-20b: 42.4 (#192)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1405 | 1269 |
| LMArena Chinese | 1427 | 1311 |
| LMArena German | 1412 | 1249 |
| LMArena Japanese | 1378 | 1245 |
| LMArena Korean | 1381 | 1239 |
| LMArena Russian | 1421 | 1279 |
| LMArena Spanish | 1448 | 1267 |
| LMArena French | 1432 | — |
Instruction Following Claude Opus 4.1 leads
Claude Opus 4.1: 75.8 (#53), gpt-oss-20b: 62.0 (#235)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1436 | 1236 |
| IFEval | — | 73.2% |
Long Context Claude Opus 4.1 leads
Claude Opus 4.1: 44.5 (#60), gpt-oss-20b: 37.9 (#203)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1455 | 1251 |
Writing & Preference Claude Opus 4.1 leads
Claude Opus 4.1: 62.6 (#70), gpt-oss-20b: 35.8 (#260)
| Benchmark | Claude Opus 4.1 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1419 | 1288 |
| LMArena Creative Writing | 1412 | 1201 |
| LMArena Multi-Turn | 1444 | 1268 |
| Short-Story Creative Writing | 84.7% | — |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
Frequently asked questions
Is Claude Opus 4.1 better than gpt-oss-20b?
Claude Opus 4.1 is the stronger model overall, scoring 40.8 to 32.1 on the Noometry Index. gpt-oss-20b costs 833× less per token, which makes it the better buy when Claude Opus 4.1's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4.1 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 Opus 4.1 lists at $15 and $75.
Is Claude Opus 4.1 or gpt-oss-20b better for coding?
Claude Opus 4.1 scores higher on coding benchmarks: 44.3 versus 37.1 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4.1 does, with 200K tokens against 131K.
How many benchmarks do Claude Opus 4.1 and gpt-oss-20b share?
25 benchmarks have published results for both models. Claude Opus 4.1 has 48 scored results on Noometry and gpt-oss-20b has 34.