Model comparison
GPT-4.1 mini vs gpt-oss-120b
gpt-oss-120b is the stronger model overall, scoring 36.0 to 33.5 on the Noometry Index.
Last verified . 37 shared benchmarks.
Summary
- They share 37 benchmarks with published results for both. GPT-4.1 mini scores higher in 4 categories and gpt-oss-120b in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 50.6 to 23.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 44.7% for GPT-4.1 mini and 88.9% for gpt-oss-120b.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.40 / $1.60 for GPT-4.1 mini.
- GPT-4.1 mini accepts more context: 1.05M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| GPT-4.1 mini | gpt-oss-120b | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 33.5 | 36.0 |
| Released | 2025-04-14 | 2025-08-05 |
| Weights | Proprietary | Open |
| Context window | 1.05M | 131K |
| Max output | 33K | 41K |
| Input $ / M tokens | $0.40 | $0.037 |
| Output $ / M tokens | $1.60 | $0.17 |
| Results tracked | 47 | 48 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-120b leads
GPT-4.1 mini: 30.4 (#286), gpt-oss-120b: 33.2 (#251)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| SWE-bench Verified (bash only) | 23.9% | 26% |
| Aider Polyglot | 32.4% | 41.8% |
| SciCode | 40.4% | 36% |
| WeirdML | 37.6% | 48.2% |
| LMArena Coding | 1367 | 1380 |
| BigCodeBench Instruct | 48.9% | — |
| CadEval | 16% | — |
| ALE-Bench | — | 575.62 |
| AlgoTune | — | 1.41 |
Agentic & Tool Use GPT-4.1 mini leads
GPT-4.1 mini: 33.0 (#52), gpt-oss-120b: 12.1 (#145)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| Terminal-Bench | — | 18.7% |
| APEX-Agents | — | 4.4% |
| Berkeley Function Calling Leaderboard | 50.5% | — |
| METR Time Horizons | — | 56.6% |
| Vending-Bench 2 | — | -21.53 |
Reasoning gpt-oss-120b leads
GPT-4.1 mini: 10.9 (#335), gpt-oss-120b: 19.5 (#236)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| Kagi LLM Benchmark | 48.6% | 58.6% |
| CritPt | 0% | 1.1% |
| Chess Puzzles | 7% | 20% |
| LMArena Hard Prompts | 1349 | 1364 |
| Mystery Game Puzzles | 7% | 2% |
| DTBench | 68.8% | 76.3% |
| LMCA | 21.1% | 22.1% |
| Epoch Capabilities Index | 135.01 | 139.94 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | — | 22.1% |
| ARC-AGI-1 | 3.5% | — |
| Surface Evolver Bench | — | 25% |
Math gpt-oss-120b leads
GPT-4.1 mini: 23.1 (#264), gpt-oss-120b: 50.6 (#43)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 44.7% | 88.9% |
| Omni-MATH | 49.1% | 68.8% |
| LMArena Math | 1343 | 1388 |
| FrontierMath (Tiers 1-3) | 6.7% | — |
| MATH Level 5 | 87.3% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
Knowledge gpt-oss-120b leads
GPT-4.1 mini: 34.7 (#183), gpt-oss-120b: 42.3 (#89)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| GPQA Diamond | 65.8% | 75.8% |
| MMLU-Pro | 78.3% | 79.5% |
| GPQA (HELM) | 61.4% | 68.4% |
| LMArena Expert | 1338 | 1356 |
| SimpleQA Verified | 12.7% | — |
| Confabulations | — | 15.7% |
| Vectara Hallucination Rate | — | 14.2% |
Multimodal Not comparable
GPT-4.1 mini: 35.2 (#77), gpt-oss-120b: —
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| LMArena Vision | 1181 | — |
Multilingual gpt-oss-120b leads
GPT-4.1 mini: 45.9 (#161), gpt-oss-120b: 48.2 (#142)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| LMArena Non-English | 1318 | 1350 |
| LMArena Chinese | 1329 | 1384 |
| LMArena French | 1357 | 1370 |
| LMArena German | 1349 | 1355 |
| LMArena Japanese | 1291 | 1332 |
| LMArena Korean | 1298 | 1283 |
| LMArena Russian | 1324 | 1343 |
| LMArena Spanish | 1319 | 1388 |
Instruction Following GPT-4.1 mini leads
GPT-4.1 mini: 73.9 (#113), gpt-oss-120b: 69.5 (#168)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| IFEval | 90.4% | 83.6% |
| LMArena Instruction Following | 1332 | 1318 |
Long Context Too close to call
GPT-4.1 mini: 31.8 (#270), gpt-oss-120b: 31.4 (#273)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| Fiction.LiveBench | 44.4% | 44.4% |
| LMArena Longer Query | 1344 | 1319 |
Writing & Preference GPT-4.1 mini leads
GPT-4.1 mini: 48.8 (#193), gpt-oss-120b: 46.7 (#213)
| Benchmark | GPT-4.1 mini | gpt-oss-120b |
|---|---|---|
| LMArena Text | 1341 | 1366 |
| LMArena Creative Writing | 1300 | 1274 |
| EQ-Bench Creative Writing | 1147 | 961 |
| WildBench | 83.8% | 84.5% |
| LMArena Multi-Turn | 1354 | 1340 |
| Short-Story Creative Writing | — | 77.1% |
Frequently asked questions
Is GPT-4.1 mini better than gpt-oss-120b?
gpt-oss-120b is the stronger model overall, scoring 36.0 to 33.5 on the Noometry Index.
Which is cheaper, GPT-4.1 mini 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-4.1 mini lists at $0.40 and $1.60.
Is GPT-4.1 mini or gpt-oss-120b better for coding?
gpt-oss-120b scores higher on coding benchmarks: 33.2 versus 30.4 in the Noometry coding category.
Which has the bigger context window?
GPT-4.1 mini does, with 1.05M tokens against 131K.
How many benchmarks do GPT-4.1 mini and gpt-oss-120b share?
37 benchmarks have published results for both models. GPT-4.1 mini has 47 scored results on Noometry and gpt-oss-120b has 48.