Model comparison
gpt-oss-20b vs Qwen3.8 Max
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 83× less per token, which makes it the better buy when Qwen3.8 Max's lead doesn't matter for your workload.
Last verified . 24 shared benchmarks.
Alibaba (Qwen)Summary
- They share 24 benchmarks with published results for both. gpt-oss-20b scores higher in 0 categories and Qwen3.8 Max in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Qwen3.8 Max leads 45.4 to 9.3.
- The biggest single-benchmark swing is Chess Puzzles: 4% for gpt-oss-20b and 40% for Qwen3.8 Max.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $2 / $6 for Qwen3.8 Max.
- Qwen3.8 Max accepts more context: 1M tokens versus 131K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-20b | Qwen3.8 Max | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 32.5 | 56.8 |
| Released | 2025-08-05 | 2026-08-02 |
| Weights | Open | Proprietary |
| Context window | 131K | 1M |
| Max output | 16K | 131K |
| Input $ / M tokens | $0.018 | $2 |
| Output $ / M tokens | $0.09 | $6 |
| Results tracked | 34 | 39 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Qwen3.8 Max leads
gpt-oss-20b: 37.6 (#192), Qwen3.8 Max: 53.5 (#29)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| SciCode | 34.4% | 53.2% |
| LMArena Coding | 1306 | 1502 |
| DeepSWE | — | 57.5% |
| LMArena WebDev | — | 1674 |
| FrontierSWE | — | 17.8% |
| WeirdML | 40.9% | — |
| ALE-Bench | 566.05 | — |
Agentic & Tool Use Qwen3.8 Max leads
gpt-oss-20b: 9.3 (#154), Qwen3.8 Max: 45.4 (#14)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| APEX-Agents | — | 63.3% |
| τ²-bench Banking | — | 55.1% |
| GDP.pdf | — | 23.2% |
Reasoning Qwen3.8 Max leads
gpt-oss-20b: 19.3 (#261), Qwen3.8 Max: 54.4 (#26)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| CritPt | 1.4% | 20% |
| Chess Puzzles | 4% | 40% |
| LMArena Hard Prompts | 1274 | 1496 |
| DTBench | 68% | 92% |
| LMCA | 14.5% | 46.2% |
| Epoch Capabilities Index | 137.82 | 156.41 |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 88.3% |
| Mystery Game Puzzles | — | 38% |
Math Qwen3.8 Max leads
gpt-oss-20b: 39.4 (#103), Qwen3.8 Max: 73.2 (#20)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 65.3% | 100% |
| LMArena Math | 1317 | 1499 |
| FrontierMath (Tiers 1-3) | — | 74.7% |
| FrontierMath Tier 4 | — | 46.3% |
| ProofBench | — | 58% |
| Omni-MATH | 56.5% | — |
Knowledge Qwen3.8 Max leads
gpt-oss-20b: 34.6 (#195), Qwen3.8 Max: 61.7 (#27)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| GPQA Diamond | 60.8% | 92.7% |
| LMArena Expert | 1258 | 1507 |
| SimpleQA Verified | — | 47.3% |
| MMLU-Pro | 74% | — |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
gpt-oss-20b: —, Qwen3.8 Max: 37.2 (#75)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| LMArena Vision | — | 1314 |
| Furniture Assembly | — | 20% |
Multilingual Qwen3.8 Max leads
gpt-oss-20b: 42.2 (#197), Qwen3.8 Max: 56.7 (#18)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| LMArena Non-English | 1268 | 1472 |
| LMArena Chinese | 1314 | 1538 |
| LMArena German | 1255 | 1483 |
| LMArena Japanese | 1244 | 1467 |
| LMArena Korean | 1236 | 1461 |
| LMArena Russian | 1278 | 1481 |
| LMArena Spanish | 1267 | 1492 |
| LMArena French | — | 1503 |
Instruction Following Qwen3.8 Max leads
gpt-oss-20b: 61.8 (#240), Qwen3.8 Max: 77.6 (#17)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| LMArena Instruction Following | 1236 | 1479 |
| IFEval | 73.2% | — |
Long Context Qwen3.8 Max leads
gpt-oss-20b: 37.9 (#209), Qwen3.8 Max: 45.6 (#31)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| LMArena Longer Query | 1250 | 1489 |
Writing & Preference Qwen3.8 Max leads
gpt-oss-20b: 35.5 (#265), Qwen3.8 Max: 67.1 (#30)
| Benchmark | gpt-oss-20b | Qwen3.8 Max |
|---|---|---|
| LMArena Text | 1287 | 1483 |
| LMArena Creative Writing | 1201 | 1479 |
| LMArena Multi-Turn | 1268 | 1489 |
| EQ-Bench Creative Writing | 666 | — |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Qwen3.8 Max?
Qwen3.8 Max is the stronger model overall, scoring 56.8 to 32.5 on the Noometry Index. gpt-oss-20b costs 83× 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-20b or Qwen3.8 Max?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Qwen3.8 Max lists at $2 and $6.
Is gpt-oss-20b or Qwen3.8 Max better for coding?
Qwen3.8 Max scores higher on coding benchmarks: 53.5 versus 37.6 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-20b and Qwen3.8 Max share?
24 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Qwen3.8 Max has 39.