Mistral 7B Instruct is a compact, 7B parameter model optimized for fast and efficient text and code generation with a 32K token context window.
Specification
most-agreed values
Context
32K
Max output
8K
Released
2025-05-26
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text, pdf
Output
text
Price
US dollars per million tokens · most-agreed
Input
$0.159
Output
$0.219
Available from 3 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| ollama · ollama | free | free | — | — | 33K | 33K | — |
| replicate · replicate | $0.05 | $0.25 | — | — | 4K | 4K | — |
| cortecs | $0.159 | $0.219 | — | — | 32K | 8K | — |
More from Mistral
most-hosted first