gpt-oss-120b (free) vs Qwen3.5 397B A17B
Lado a lado. Cada métrica. Cada benchmark.
| Tipo | gpt-oss-120b (free) | Qwen3.5 397B A17B |
|---|---|---|
| Provider |  OpenAI |  Alibaba Qwen |
| pontuação média | 68.7 | 78.4 |
| Preço de entrada | $0.00 | $0.39 |
| Preço de saída | $0.00 | $2.34 |
| Janela de contexto | 131K tokens (~66 books) | 262K tokens (~131 books) |
| Lançado em | 2025-08-05 | 2026-02-16 |
| Código aberto | Open Source | Open Source |
Pontuações de benchmark
9 benchmarks · gpt-oss-120b (free): 1, Qwen3.5 397B A17B: 8
| Benchmark | Categoria | gpt-oss-120b (free) | Qwen3.5 397B A17B |
|---|---|---|---|
| Artificial Analysis — Agentic Index | speed | 37.9 | 55.8 |
| Artificial Analysis — Coding Index | speed | 28.6 | 41.3 |
| Artificial Analysis — Quality Index | speed | 33.3 | 45.0 |
| OpenCompass — AIME2025 | math | 93.4 | 92.3 |
| OpenCompass — GPQA-Diamond | knowledge | 78.9 | 88.4 |
| OpenCompass — HLE | knowledge | 18.3 | 27.5 |
| OpenCompass — IFEval | language | 90.2 | 91.5 |
| OpenCompass — LiveCodeBenchV6 | coding | 78.4 | 83.0 |
| OpenCompass — MMLU-Pro | knowledge | 79.7 | 87.6 |