gpt-oss-20b is 1.9× cheaper per million tokens (blended). gpt-oss-20b has the larger context window (131,072 vs 32,768 tokens, 4.0×).
gpt-oss-20b vs Qwen2.5 7B Instruct
Live catalog fields. Best-in-row is highlighted.
gpt-oss-20bopen
OpenAI · openai__gpt-oss-20b
Qwen2.5 7B Instructopen
Qwen · qwen__qwen-2.5-7b-instruct
Identity
| Field | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
|---|---|---|
| Provider | OpenAI | Qwen |
| Slug | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
| Status | live | live |
| Open weights | Yes | Yes |
| License | — | — |
| HuggingFace | openai/gpt-oss-20b | Qwen/Qwen2.5-7B-Instruct |
Quality
| Field | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
|---|---|---|
| Input $ / 1M tokens | $0.03 | $0.10 |
| Cached input $ / 1M | $0.03 | — |
| Output $ / 1M tokens | $0.13 | $0.20 |
Context
| Field | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
|---|---|---|
| Max context tokens | 131,072 | 32,768 |
| Max output tokens | 117,964 | 29,491 |
Modalities
| Field | openai__gpt-oss-20b | qwen__qwen-2.5-7b-instruct |
|---|---|---|
| Input | text | text |
| Output | text | text |