Model comparison
GPT-5.1 vs gpt-oss-120b
GPT-5.1 is the stronger model overall, scoring 49.0 to 36.3 on the Noometry Index. gpt-oss-120b costs 49× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Last verified . 38 shared benchmarks.
Summary
- They share 38 benchmarks with published results for both. GPT-5.1 scores higher in 8 categories and gpt-oss-120b in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where GPT-5.1 leads 32.7 to 12.2.
- The biggest single-benchmark swing is SWE-bench Verified (bash only): 66% for GPT-5.1 and 26% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $1.25 / $10 for GPT-5.1.
- GPT-5.1 accepts more context: 400K tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| GPT-5.1 | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 49.0 | 36.3 |
| Released | 2025-11-13 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 41K |
| Input $ / M tokens | $1.25 | $0.037 |
| Output $ / M tokens | $10 | $0.17 |
| Results tracked | 63 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.1 leads
GPT-5.1: 46.4 (#66), gpt-oss-120b: 33.5 (#256)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| SWE-bench Verified (bash only) | 66% | 26% |
| SciCode | 43.3% | 36% |
| WeirdML | 60.8% | 48.2% |
| LMArena Coding | 1454 | 1380 |
| ALE-Bench | 1,192 | 575.62 |
| SWE-bench Verified | 68% | — |
| Aider Polyglot | — | 41.8% |
| LMArena WebDev | 1395 | — |
| GSO | 13.7% | — |
| LiveBench Coding | 72.5% | — |
| AlgoTune | — | 1.41 |
Agentic & Tool Use GPT-5.1 leads
GPT-5.1: 32.7 (#60), gpt-oss-120b: 12.2 (#153)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | 47.6% | 18.7% |
| Vending-Bench 2 | 1,473 | -21.53 |
| APEX-Agents | — | 4.4% |
| DeepResearch Bench | 42.8% | — |
| LMArena Search | 1199 | — |
| METR Time Horizons | — | 56.6% |
Reasoning GPT-5.1 leads
GPT-5.1: 39.8 (#58), gpt-oss-120b: 20.0 (#245)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| SimpleBench | 53.2% | 22.1% |
| CritPt | 4.9% | 1.1% |
| Chess Puzzles | 32% | 20% |
| LMArena Hard Prompts | 1457 | 1364 |
| Mystery Game Puzzles | 19% | 2% |
| DTBench | 90.1% | 76.3% |
| LMCA | 43.9% | 22.1% |
| Epoch Capabilities Index | 149.64 | 139.93 |
| ARC-AGI-2 | 17.6% | — |
| Kagi LLM Benchmark | — | 58.6% |
| ARC-AGI-1 | 72.8% | — |
| EnigmaEval | 11.2% | — |
| LiveBench Reasoning | 95.8% | — |
| LiveBench Data Analysis | 72.1% | — |
| Surface Evolver Bench | — | 25% |
| ForecastBench | 58.1 | — |
| LiveBench | 78.8% | — |
Math Too close to call
GPT-5.1: 52.2 (#51), gpt-oss-120b: 52.5 (#50)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.6% | 88.9% |
| Omni-MATH | 46.4% | 68.8% |
| LMArena Math | 1447 | 1389 |
| LiveBench Math | 94.5% | — |
| FrontierMath (Feb 2025 set) | 31% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |
Knowledge GPT-5.1 leads
GPT-5.1: 50.6 (#71), gpt-oss-120b: 42.4 (#96)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 87.6% | 75.8% |
| MMLU-Pro | 57.9% | 79.5% |
| Vectara Hallucination Rate | 10.9% | 14.2% |
| GPQA (HELM) | 44.2% | 68.4% |
| LMArena Expert | 1470 | 1356 |
| Humanity's Last Exam | 23.7% | — |
| SimpleQA Verified | 48% | — |
| Confabulations | — | 15.7% |
Multimodal Not comparable
GPT-5.1: 44.8 (#19), gpt-oss-120b: —
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1250 | — |
| VPCT | 58.7% | — |
| LMArena Document | 1403 | — |
Multilingual GPT-5.1 leads
GPT-5.1: 53.8 (#56), gpt-oss-120b: 48.0 (#147)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1431 | 1351 |
| LMArena Chinese | 1495 | 1385 |
| LMArena French | 1450 | 1369 |
| LMArena German | 1438 | 1353 |
| LMArena Japanese | 1453 | 1331 |
| LMArena Korean | 1401 | 1282 |
| LMArena Russian | 1435 | 1343 |
| LMArena Spanish | 1433 | 1389 |
Instruction Following GPT-5.1 leads
GPT-5.1: 83.9 (#1), gpt-oss-120b: 69.3 (#173)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| IFEval | 93.5% | 83.6% |
| LMArena Instruction Following | 1443 | 1318 |
| LiveBench Instruction Following | 93.3% | — |
Long Context GPT-5.1 leads
GPT-5.1: 47.6 (#14), gpt-oss-120b: 31.4 (#278)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| LMArena Longer Query | 1447 | 1319 |
| Fiction.LiveBench | — | 44.4% |
| CL-bench | 23.7% | — |
| CL-bench Life | 17.3% | — |
Writing & Preference GPT-5.1 leads
GPT-5.1: 64.5 (#55), gpt-oss-120b: 46.5 (#217)
| Benchmark | GPT-5.1 | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1443 | 1365 |
| LMArena Creative Writing | 1427 | 1275 |
| WildBench | 86.3% | 84.5% |
| LMArena Multi-Turn | 1450 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
| EQ-Bench Creative Writing | — | 961 |
| LiveBench Language | 80.2% | — |
Frequently asked questions
Is GPT-5.1 better than gpt-oss-120b?
GPT-5.1 is the stronger model overall, scoring 49.0 to 36.3 on the Noometry Index. gpt-oss-120b costs 49× less per token, which makes it the better buy when GPT-5.1's lead doesn't matter for your workload.
Which is cheaper, GPT-5.1 or gpt-oss-120b?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; GPT-5.1 lists at $1.25 and $10.
Is GPT-5.1 or gpt-oss-120b better for coding?
GPT-5.1 scores higher on coding benchmarks: 46.4 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
GPT-5.1 does, with 400K tokens against 131K.
How many benchmarks do GPT-5.1 and gpt-oss-120b share?
38 benchmarks have published results for both models. GPT-5.1 has 63 scored results on Noometry and gpt-oss-120b has 48.