Model comparison
gpt-oss-120b vs Qwen2.5-Coder-32B
gpt-oss-120b is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Alibaba (Qwen)Summary
- They share 15 benchmarks with published results for both. gpt-oss-120b scores higher in 6 categories and Qwen2.5-Coder-32B in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 33.3.
- The biggest single-benchmark swing is Aider Polyglot: 41.8% for gpt-oss-120b and 16.4% for Qwen2.5-Coder-32B.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $0.66 / $1 for Qwen2.5-Coder-32B.
- gpt-oss-120b accepts more context: 131K tokens versus 33K.
Side by side
| gpt-oss-120b | Qwen2.5-Coder-32B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 33.4 |
| Released | 2025-08-05 | 2024-09-18 |
| Weights | Open | Open |
| Context window | 131K | 33K |
| Max output | 41K | 29K |
| Input $ / M tokens | $0.037 | $0.66 |
| Output $ / M tokens | $0.17 | $1 |
| Results tracked | 48 | 31 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-120b leads
gpt-oss-120b: 33.5 (#256), Qwen2.5-Coder-32B: 22.6 (#333)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| SWE-bench Verified (bash only) | 26% | 9% |
| Aider Polyglot | 41.8% | 16.4% |
| LMArena Coding | 1380 | 1276 |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| BigCodeBench Instruct | — | 49% |
| LiveBench Coding | — | 56.9% |
| BigCodeBench Complete | — | 58% |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 77% |
Agentic & Tool Use Not comparable
gpt-oss-120b: 12.2 (#153), Qwen2.5-Coder-32B: —
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Qwen2.5-Coder-32B leads
gpt-oss-120b: 20.0 (#245), Qwen2.5-Coder-32B: 21.2 (#225)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1251 |
| Epoch Capabilities Index | 139.93 | 119.49 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| LiveBench Reasoning | — | 42.1% |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LiveBench Data Analysis | — | 49.9% |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| HellaSwag | — | 83% |
| LiveBench | — | 46.2% |
| WinoGrande | — | 80.8% |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Qwen2.5-Coder-32B: 33.3 (#204)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Math | 1389 | 1251 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
| LiveBench Math | — | 46.6% |
| GSM8K | — | 93% |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Qwen2.5-Coder-32B: 33.4 (#203)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Expert | 1356 | 1221 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
| ARC (AI2) Challenge | — | 70.5% |
| MMLU | — | 79.1% |
Multilingual gpt-oss-120b leads
gpt-oss-120b: 48.0 (#147), Qwen2.5-Coder-32B: 37.8 (#235)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Non-English | 1351 | 1205 |
| LMArena Chinese | 1385 | 1222 |
| LMArena Russian | 1343 | 1228 |
| LMArena French | 1369 | — |
| LMArena German | 1353 | — |
| LMArena Japanese | 1331 | — |
| LMArena Korean | 1282 | — |
| LMArena Spanish | 1389 | — |
Instruction Following gpt-oss-120b leads
gpt-oss-120b: 69.3 (#173), Qwen2.5-Coder-32B: 61.4 (#245)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Instruction Following | 1318 | 1223 |
| LiveBench Instruction Following | — | 58.7% |
| IFEval | 83.6% | — |
Long Context Qwen2.5-Coder-32B leads
gpt-oss-120b: 31.4 (#278), Qwen2.5-Coder-32B: 38.0 (#208)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Longer Query | 1319 | 1251 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference gpt-oss-120b leads
gpt-oss-120b: 46.5 (#217), Qwen2.5-Coder-32B: 41.6 (#240)
| Benchmark | gpt-oss-120b | Qwen2.5-Coder-32B |
|---|---|---|
| LMArena Text | 1365 | 1230 |
| LMArena Creative Writing | 1275 | 1174 |
| LMArena Multi-Turn | 1340 | 1222 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| LiveBench Language | — | 23.3% |
Frequently asked questions
Is gpt-oss-120b better than Qwen2.5-Coder-32B?
gpt-oss-120b is the stronger model overall, scoring 36.3 to 33.4 on the Noometry Index.
Which is cheaper, gpt-oss-120b or Qwen2.5-Coder-32B?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Qwen2.5-Coder-32B lists at $0.66 and $1.
Is gpt-oss-120b or Qwen2.5-Coder-32B better for coding?
gpt-oss-120b scores higher on coding benchmarks: 33.5 versus 22.6 in the Noometry coding category.
Which has the bigger context window?
gpt-oss-120b does, with 131K tokens against 33K.
How many benchmarks do gpt-oss-120b and Qwen2.5-Coder-32B share?
15 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Qwen2.5-Coder-32B has 31.