Mistral: Saba
mistralai/mistral-saba
Saba is Mistral's budget option at $0.2/$0.6 per million input/output tokens, with a 32K-token context. It sits toward the back of the board on capability, at #232 of 234 on the TryAii Score — its best showing is #5 on HumanEval. It's among the fastest models we track, at about 122 tokens/sec.
Prompt $/1M
$0.200
Completion $/1M
$0.600
Context
32,768 tok
Best provider
123 tok/s · Mistral
Specs
- Model id
- mistralai/mistral-saba
- Display name
- Mistral: Saba
- Modality
- text+file->text
- Tokenizer
- Mistral
- Instruct type
- —
- Context length
- 32,768
- Max completion
- 26,214
- Prompt $/1M
- $0.200
- Completion $/1M
- $0.600
- Free tier
- no
Capabilities
- Function callingno
- JSON modeno
- System promptno
- Visionno
- Moderatedno
Benchmark scores
Each axis is one benchmark, normalized to the best model in the field.
No benchmark data for this model.
Speed by provider
| Provider | tok/s |
|---|---|
| Mistralfastest | 123 |
| mistralaifastest | — |
| Artificial Analysis | 0 |