Model comparison
gpt-oss-20b vs o1-mini
o1-mini is the stronger model overall, scoring 34.0 to 32.5 on the Noometry Index.
Last verified . 20 shared benchmarks.
Summary
- They share 20 benchmarks with published results for both. gpt-oss-20b scores higher in 3 categories and o1-mini in 6 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where o1-mini leads 24.6 to 9.3.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 65.3% for gpt-oss-20b and 46.9% for o1-mini.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-20b | o1-mini | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 32.5 | 34.0 |
| Released | 2025-08-05 | 2024-09-12 |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.018 | — |
| Output $ / M tokens | $0.09 | — |
| Results tracked | 34 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.6 (#192), o1-mini: 35.5 (#224)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| WeirdML | 40.9% | 36.3% |
| LMArena Coding | 1306 | 1362 |
| Aider Polyglot | — | 32.9% |
| SciCode | 34.4% | — |
| LiveBench Coding | — | 48% |
| ALE-Bench | 566.05 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |
Agentic & Tool Use o1-mini leads
gpt-oss-20b: 9.3 (#154), o1-mini: 24.6 (#118)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| Cybench | — | 10% |
Reasoning gpt-oss-20b leads
gpt-oss-20b: 19.3 (#261), o1-mini: 8.8 (#346)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1333 |
| Epoch Capabilities Index | 137.82 | 135.82 |
| ARC-AGI-2 | — | 0.8% |
| SimpleBench | — | 18.1% |
| Kagi LLM Benchmark | 53.2% | — |
| ARC-AGI-1 | — | 14% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| LiveBench Reasoning | — | 72.3% |
| DTBench | 68% | — |
| LiveBench Data Analysis | — | 57.9% |
| LMCA | 14.5% | — |
| LiveBench | — | 57.8% |
Math gpt-oss-20b leads
gpt-oss-20b: 39.4 (#103), o1-mini: 35.4 (#186)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 46.9% |
| LMArena Math | 1317 | 1358 |
| Omni-MATH | 56.5% | — |
| LiveBench Math | — | 62% |
| MATH Level 5 | — | 89.2% |
| FrontierMath (Feb 2025 set) | — | 1.7% |
Knowledge Too close to call
gpt-oss-20b: 34.6 (#195), o1-mini: 34.9 (#192)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| GPQA Diamond | 60.8% | 62.4% |
| LMArena Expert | 1258 | 1316 |
| MMLU-Pro | 74% | — |
| Confabulations | — | 18.6% |
| GPQA (HELM) | 59.4% | — |
Multilingual o1-mini leads
gpt-oss-20b: 42.2 (#197), o1-mini: 43.6 (#182)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| LMArena Non-English | 1268 | 1289 |
| LMArena Chinese | 1314 | 1314 |
| LMArena German | 1255 | 1278 |
| LMArena Japanese | 1244 | 1245 |
| LMArena Korean | 1236 | 1223 |
| LMArena Russian | 1278 | 1283 |
| LMArena Spanish | 1267 | 1303 |
| LMArena French | — | 1293 |
Instruction Following o1-mini leads
gpt-oss-20b: 61.8 (#240), o1-mini: 66.7 (#206)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1236 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
| IFEval | 73.2% | — |
Long Context o1-mini leads
gpt-oss-20b: 37.9 (#209), o1-mini: 40.1 (#161)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| LMArena Longer Query | 1250 | 1320 |
Writing & Preference o1-mini leads
gpt-oss-20b: 35.5 (#265), o1-mini: 48.4 (#202)
| Benchmark | gpt-oss-20b | o1-mini |
|---|---|---|
| LMArena Text | 1287 | 1317 |
| LMArena Creative Writing | 1201 | 1244 |
| LMArena Multi-Turn | 1268 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
| LiveBench Language | — | 40.9% |
Frequently asked questions
Is gpt-oss-20b better than o1-mini?
o1-mini is the stronger model overall, scoring 34.0 to 32.5 on the Noometry Index.
Is gpt-oss-20b or o1-mini better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.6 versus 35.5 in the Noometry coding category.
How many benchmarks do gpt-oss-20b and o1-mini share?
20 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and o1-mini has 39.