Model comparison
gpt-oss-120b vs Qwen2.5-Max
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 36.3 on the Noometry Index.
Last verified . 20 shared benchmarks.
Alibaba (Qwen)Summary
- They share 20 benchmarks with published results for both. gpt-oss-120b scores higher in 2 categories and Qwen2.5-Max in 6 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where gpt-oss-120b leads 52.5 to 36.9.
- The biggest single-benchmark swing is Confabulations: 15.7% for gpt-oss-120b and 21.8% for Qwen2.5-Max.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-120b | Qwen2.5-Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 40.7 |
| Released | 2025-08-05 | 2025-01-25 |
| Weights | Open | Proprietary |
| Context window | 131K | — |
| Max output | 41K | — |
| Input $ / M tokens | $0.037 | — |
| Output $ / M tokens | $0.17 | — |
| Results tracked | 48 | 27 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen2.5-Max leads
gpt-oss-120b: 33.5 (#256), Qwen2.5-Max: 41.8 (#117)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1380 | 1359 |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| SciCode | 36% | — |
| WeirdML | 48.2% | — |
| LiveBench Coding | — | 64.4% |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Not comparable
gpt-oss-120b: 12.2 (#153), Qwen2.5-Max: —
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| Terminal-Bench | 18.7% | — |
| APEX-Agents | 4.4% | — |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Qwen2.5-Max leads
gpt-oss-120b: 20.0 (#245), Qwen2.5-Max: 25.6 (#147)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1360 |
| Epoch Capabilities Index | 139.93 | 132.53 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| CritPt | 1.1% | — |
| Chess Puzzles | 20% | — |
| LiveBench Reasoning | — | 51.4% |
| Mystery Game Puzzles | 2% | — |
| DTBench | 76.3% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 22.1% | — |
| Surface Evolver Bench | 25% | — |
| LiveBench | — | 62.3% |
Math gpt-oss-120b leads
gpt-oss-120b: 52.5 (#50), Qwen2.5-Max: 36.9 (#162)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1389 | 1369 |
| OTIS Mock AIME 2024-2025 | 88.9% | — |
| Omni-MATH | 68.8% | — |
| LiveBench Math | — | 58.4% |
Knowledge gpt-oss-120b leads
gpt-oss-120b: 42.4 (#96), Qwen2.5-Max: 35.3 (#186)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| Confabulations | 15.7% | 21.8% |
| LMArena Expert | 1356 | 1337 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 79.5% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multilingual Too close to call
gpt-oss-120b: 48.0 (#147), Qwen2.5-Max: 48.1 (#146)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1351 | 1352 |
| LMArena Chinese | 1385 | 1382 |
| LMArena French | 1369 | 1396 |
| LMArena German | 1353 | 1350 |
| LMArena Japanese | 1331 | 1300 |
| LMArena Korean | 1282 | 1304 |
| LMArena Russian | 1343 | 1353 |
| LMArena Spanish | 1389 | 1377 |
Instruction Following Qwen2.5-Max leads
gpt-oss-120b: 69.3 (#173), Qwen2.5-Max: 71.3 (#152)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1318 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 83.6% | — |
Long Context Qwen2.5-Max leads
gpt-oss-120b: 31.4 (#278), Qwen2.5-Max: 41.4 (#142)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1319 | 1358 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Qwen2.5-Max leads
gpt-oss-120b: 46.5 (#217), Qwen2.5-Max: 55.4 (#146)
| Benchmark | gpt-oss-120b | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1365 | 1367 |
| LMArena Creative Writing | 1275 | 1339 |
| Short-Story Creative Writing | 77.1% | 72.9% |
| LMArena Multi-Turn | 1340 | 1364 |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
| LiveBench Language | — | 56.3% |
Frequently asked questions
Is gpt-oss-120b better than Qwen2.5-Max?
Qwen2.5-Max is the stronger model overall, scoring 40.7 to 36.3 on the Noometry Index.
Is gpt-oss-120b or Qwen2.5-Max better for coding?
Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 33.5 in the Noometry coding category.
How many benchmarks do gpt-oss-120b and Qwen2.5-Max share?
20 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Qwen2.5-Max has 27.