Model comparison
gpt-oss-120b vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 43× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 27 shared benchmarks.
Alibaba (Qwen)Summary
- They share 27 benchmarks with published results for both. gpt-oss-120b scores higher in 0 categories and Qwen3.8 Max in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Qwen3.8 Max leads 54.4 to 20.0.
- The biggest single-benchmark swing is APEX-Agents: 4.4% for gpt-oss-120b and 63.3% for Qwen3.8 Max.
- gpt-oss-120b is cheaper at $0.037 / $0.17 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 131K.
- gpt-oss-120b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-120b | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.3 | 56.8 |
| Released | 2025-08-05 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 41K | 131K |
| Input $ / M tokens | $0.037 | $2 |
| Output $ / M tokens | $0.17 | $6 |
| Results tracked | 48 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
gpt-oss-120b: 33.5 (#256), Qwen3.8 Max: 53.5 (#29)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| SciCode | 36% | 53.2% |
| LMArena Coding | 1380 | 1502 |
| DeepSWE | — | 57.5% |
| SWE-bench Verified (bash only) | 26% | — |
| Aider Polyglot | 41.8% | — |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| WeirdML | 48.2% | — |
| ALE-Bench | 575.62 | — |
| AlgoTune | 1.41 | — |
Agentic & Tool Use Qwen3.8 Max leads
gpt-oss-120b: 12.2 (#153), Qwen3.8 Max: 45.4 (#14)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| APEX-Agents | 4.4% | 63.3% |
| Terminal-Bench | 18.7% | — |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
| METR Time Horizons | 56.6% | — |
| Vending-Bench 2 | -21.53 | — |
Reasoning Qwen3.8 Max leads
gpt-oss-120b: 20.0 (#245), Qwen3.8 Max: 54.4 (#26)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| CritPt | 1.1% | 20% |
| Chess Puzzles | 20% | 40% |
| LMArena Hard Prompts | 1364 | 1496 |
| Mystery Game Puzzles | 2% | 38% |
| DTBench | 76.3% | 92% |
| LMCA | 22.1% | 46.2% |
| Epoch Capabilities Index | 139.93 | 156.41 |
| SimpleBench | 22.1% | — |
| Kagi LLM Benchmark | 58.6% | — |
| NYT Connections (extended) | — | 88.3% |
| Surface Evolver Bench | 25% | — |
Math Qwen3.8 Max leads
gpt-oss-120b: 52.5 (#50), Qwen3.8 Max: 73.2 (#20)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 88.9% | 100% |
| LMArena Math | 1389 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| Omni-MATH | 68.8% | — |
Knowledge Qwen3.8 Max leads
gpt-oss-120b: 42.4 (#96), Qwen3.8 Max: 61.7 (#27)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 75.8% | 92.7% |
| LMArena Expert | 1356 | 1507 |
| SimpleQA Verified | — | 47.3% |
| MMLU-Pro | 79.5% | — |
| Confabulations | 15.7% | — |
| Vectara Hallucination Rate | 14.2% | — |
| GPQA (HELM) | 68.4% | — |
Multimodal Not comparable
gpt-oss-120b: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
gpt-oss-120b: 48.0 (#147), Qwen3.8 Max: 56.7 (#18)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1351 | 1472 |
| LMArena Chinese | 1385 | 1538 |
| LMArena French | 1369 | 1503 |
| LMArena German | 1353 | 1483 |
| LMArena Japanese | 1331 | 1467 |
| LMArena Korean | 1282 | 1461 |
| LMArena Russian | 1343 | 1481 |
| LMArena Spanish | 1389 | 1492 |
Instruction Following Qwen3.8 Max leads
gpt-oss-120b: 69.3 (#173), Qwen3.8 Max: 77.6 (#17)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1318 | 1479 |
| IFEval | 83.6% | — |
Long Context Qwen3.8 Max leads
gpt-oss-120b: 31.4 (#278), Qwen3.8 Max: 45.6 (#31)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1319 | 1489 |
| Fiction.LiveBench | 44.4% | — |
Writing & Preference Qwen3.8 Max leads
gpt-oss-120b: 46.5 (#217), Qwen3.8 Max: 67.1 (#30)
| Benchmark | gpt-oss-120b | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1365 | 1483 |
| LMArena Creative Writing | 1275 | 1479 |
| LMArena Multi-Turn | 1340 | 1489 |
| Short-Story Creative Writing | 77.1% | — |
| EQ-Bench Creative Writing | 961 | — |
| WildBench | 84.5% | — |
Frequently asked questions
Is gpt-oss-120b better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 36.3 on the Noometry Index. gpt-oss-120b costs 43× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-120b or Qwen3.8 Max?
gpt-oss-120b is cheaper. It lists at $0.037 per million input tokens and $0.17 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is gpt-oss-120b or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 33.5 in the Noometry coding category.
Which has the bigger context window?
Qwen3.8 Max does, with 1M tokens against 131K.
How many benchmarks do gpt-oss-120b and Qwen3.8 Max share?
27 benchmarks have published results for both models. gpt-oss-120b has 48 scored results on Noometry and Qwen3.8 Max has 39.