gpt-oss-120b is 7.4× cheaper per million tokens (blended). gpt-oss-120b has the larger context window (131,072 vs 32,768 tokens, 4.0×). gpt-oss-120b is open-weights; the other is not.
Saba vs gpt-oss-120b
Live catalog fields. Best-in-row is highlighted.
Saba
Mistral · mistralai__mistral-saba
gpt-oss-120bopen
OpenAI · openai__gpt-oss-120b
Identity
| Field | mistralai__mistral-saba | openai__gpt-oss-120b |
|---|---|---|
| Provider | Mistral | OpenAI |
| Slug | mistralai__mistral-saba | openai__gpt-oss-120b |
| Status | live | live |
| Open weights | No | Yes |
| License | — | — |
| HuggingFace | — | openai/gpt-oss-120b |
Quality
| Field | mistralai__mistral-saba | openai__gpt-oss-120b |
|---|---|---|
| Quality index (0–100) | — | — |
Cost
| Field | mistralai__mistral-saba | openai__gpt-oss-120b |
|---|---|---|
| Input $ / 1M tokens | $0.20 | $0.02 |
| Cached input $ / 1M | $0.02 | $0.02 |
| Output $ / 1M tokens | $0.60 | $0.09 |
Context
| Field | mistralai__mistral-saba | openai__gpt-oss-120b |
|---|---|---|
| Max context tokens | 32,768 | 131,072 |
| Max output tokens | 26,214 | 65,536 |
Modalities
| Field | mistralai__mistral-saba | openai__gpt-oss-120b |
|---|---|---|
| Input | text, file | text |
| Output | text | text |