Model comparison
gpt-oss-120b vs Qwen3-4B
gpt-oss-120b is the stronger model overall, scoring 36.0 to 31.5 on the Noometry Index.
Last verified . 4 shared benchmarks.
Alibaba (Qwen)Summary
- They share 4 benchmarks with published results for both. gpt-oss-120b scores higher in 3 categories and Qwen3-4B in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 50.6 to 28.5.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for gpt-oss-120b and 52.2% for Qwen3-4B.
Side by side
| gpt-oss-120b | Qwen3-4B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.0 | 31.5 |
| Released | 2025-08-05 | 2025-04-29 |
| Weights | Open | Open |
| Context window | 131K | — |
| Max output | 41K | — |
| Input $ / M tokens | $0.037 | — |
| Output $ / M tokens | $0.17 | — |
| Results tracked | 48 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
gpt-oss-120b: 33.2 (#251), Qwen3-4B: —
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| LMArena Coding | 1380 | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Qwen3-4B leads
gpt-oss-120b: 12.1 (#145), Qwen3-4B: 27.3 (#94)
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| Berkeley Function Calling Leaderboard | — | 35.7% |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Too close to call
gpt-oss-120b: 19.5 (#236), Qwen3-4B: 18.5 (#264)
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| Chess Puzzles | 20% | 4% |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| CritPt | 1.1% | — |
| LMArena Hard Prompts | 1364 | — |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| Epoch Capabilities Index | 139.94 | — |
Math gpt-oss-120b leads
gpt-oss-120b: 50.6 (#43), Qwen3-4B: 28.5 (#228)
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 52.2% |
| MathArena Final-Answer Competitions | — | 38.5% |
| Omni-MATH | 68.8% | — |
| LMArena Math | 1388 | — |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.3 (#89), Qwen3-4B: 33.0 (#198)
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| GPQA Diamond | 75.8% | 52.3% |
| Vectara Hallucination Rate | 14.2% | 5.7% |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| GPQA (HELM) | 68.4% | — |
| LMArena Expert | 1356 | — |
Multilingual Not comparable
gpt-oss-120b: 48.2 (#142), Qwen3-4B: —
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| LMArena Non-English | 1350 | — |
| LMArena Chinese | 1384 | — |
| LMArena French | 1370 | — |
| LMArena German | 1355 | — |
| LMArena Japanese | 1332 | — |
| LMArena Korean | 1283 | — |
| LMArena Russian | 1343 | — |
| LMArena Spanish | 1388 | — |
Instruction Following Not comparable
gpt-oss-120b: 69.5 (#168), Qwen3-4B: —
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| IFEval | 83.6% | — |
| LMArena Instruction Following | 1318 | — |
Long Context Not comparable
gpt-oss-120b: 31.4 (#273), Qwen3-4B: —
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| Fiction.LiveBench | 44.4% | — |
| LMArena Longer Query | 1319 | — |
Writing & Preference Not comparable
gpt-oss-120b: 46.7 (#213), Qwen3-4B: —
| Benchmark | gpt-oss-120b | Qwen3-4B |
|---|---|---|
| LMArena Text | 1366 | — |
| LMArena Creative Writing | 1274 | — |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| LMArena Multi-Turn | 1340 | — |
Frequently asked questions
Is gpt-oss-120b better than Qwen3-4B?
gpt-oss-120b is the stronger model overall, scoring 36.0 to 31.5 on the Noometry Index.
How many benchmarks do gpt-oss-120b and Qwen3-4B share?
4 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Qwen3-4B has 6.