Model comparison
Claude Opus 4 vs gpt-oss-20b
Claude Opus 4 is the stronger model overall, scoring 43.1 to 32.5 on the Noometry Index. gpt-oss-20b costs 833× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. Claude Opus 4 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 Opus 4 leads 61.2 to 35.5.
- The biggest single-benchmark swing is LMCA: 37.4% for Claude Opus 4 and 14.5% 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.
- Claude Opus 4 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 | gpt-oss-20b | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 32.5 |
| Released | 2025-05-22 | 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 | 56 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Opus 4 leads
Claude Opus 4: 47.2 (#62), gpt-oss-20b: 37.6 (#192)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| WeirdML | 43.7% | 40.9% |
| LMArena Coding | 1442 | 1306 |
| SWE-bench Verified | 70.7% | — |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| SciCode | — | 34.4% |
| GSO | 6.9% | — |
| ALE-Bench | — | 566.05 |
| AlgoTune | 1.33 | — |
Agentic & Tool Use Claude Opus 4 leads
Claude Opus 4: 34.8 (#42), gpt-oss-20b: 9.3 (#154)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |
Reasoning Claude Opus 4 leads
Claude Opus 4: 27.3 (#121), gpt-oss-20b: 19.3 (#261)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| Kagi LLM Benchmark | 74.3% | 53.2% |
| CritPt | 0.3% | 1.4% |
| LMArena Hard Prompts | 1399 | 1274 |
| DTBench | 81.6% | 68% |
| LMCA | 37.4% | 14.5% |
| Epoch Capabilities Index | 142.67 | 137.82 |
| ARC-AGI-2 | 8.6% | — |
| SimpleBench | 58.8% | — |
| ARC-AGI-1 | 35.7% | — |
| Chess Puzzles | — | 4% |
| EnigmaEval | 5.6% | — |
| ForecastBench | 61.1 | — |
Math Claude Opus 4 leads
Claude Opus 4: 42.0 (#86), gpt-oss-20b: 39.4 (#103)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 65.3% |
| Omni-MATH | 61.6% | 56.5% |
| LMArena Math | 1390 | 1317 |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |
Knowledge Claude Opus 4 leads
Claude Opus 4: 44.0 (#88), gpt-oss-20b: 34.6 (#195)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | 76.3% | 60.8% |
| MMLU-Pro | 87.5% | 74% |
| GPQA (HELM) | 70.8% | 59.4% |
| LMArena Expert | 1386 | 1258 |
| Humanity's Last Exam | 10.7% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
Multimodal Not comparable
Claude Opus 4: 31.5 (#106), gpt-oss-20b: —
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |
Multilingual Claude Opus 4 leads
Claude Opus 4: 48.8 (#138), gpt-oss-20b: 42.2 (#197)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 1362 | 1268 |
| LMArena Chinese | 1386 | 1314 |
| LMArena German | 1391 | 1255 |
| LMArena Japanese | 1331 | 1244 |
| LMArena Korean | 1321 | 1236 |
| LMArena Russian | 1392 | 1278 |
| LMArena Spanish | 1389 | 1267 |
| LMArena French | 1372 | — |
Instruction Following Claude Opus 4 leads
Claude Opus 4: 77.1 (#28), gpt-oss-20b: 61.8 (#240)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| IFEval | 91.8% | 73.2% |
| LMArena Instruction Following | 1406 | 1236 |
Long Context Claude Opus 4 leads
Claude Opus 4: 39.6 (#172), gpt-oss-20b: 37.9 (#209)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | 1422 | 1250 |
| Fiction.LiveBench | 61.1% | — |
Writing & Preference Claude Opus 4 leads
Claude Opus 4: 61.2 (#89), gpt-oss-20b: 35.5 (#265)
| Benchmark | Claude Opus 4 | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1377 | 1287 |
| LMArena Creative Writing | 1387 | 1201 |
| EQ-Bench Creative Writing | 1580 | 666 |
| WildBench | 85.2% | 73.7% |
| LMArena Multi-Turn | 1396 | 1268 |
| Short-Story Creative Writing | 83.6% | — |
Frequently asked questions
Is Claude Opus 4 better than gpt-oss-20b?
Claude Opus 4 is the stronger model overall, scoring 43.1 to 32.5 on the Noometry Index. gpt-oss-20b costs 833× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.
Which is cheaper, Claude Opus 4 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 lists at $15 and $75.
Is Claude Opus 4 or gpt-oss-20b better for coding?
Claude Opus 4 scores higher on coding benchmarks: 47.2 versus 37.6 in the Noometry coding category.
Which has the bigger context window?
Claude Opus 4 does, with 200K tokens against 131K.
How many benchmarks do Claude Opus 4 and gpt-oss-20b share?
30 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and gpt-oss-20b has 34.