Magistral Medium
Complex multi-step enterprise reasoning where precision matters.
What are Magistral Medium's specs and price?
Magistral Medium, built by Mistral, ships a 128K-token context window and a 33K-token max output, released 2026-09. It supports text input with a dedicated reasoning mode and costs $2.75 per million blended tokens, the 31st-cheapest of 49 models we track.
What are Magistral Medium's specs?
| Context window | 128K tokens |
| Max output | 33K tokens |
| Modalities | text |
| Extended thinking | Yes |
| Released | 2026-09 |
| Knowledge cutoff | Not published |
| Provider | Mistral |
Verified 2026-10-01 — source.
Where does Magistral Medium rank?
What are Magistral Medium's strengths?
- Mistral enterprise reasoning model
- 128K context for long multi-step work
- Strong logic at $2/$5 per million tokens
What else should you know about Magistral Medium?
What are common questions about Magistral Medium?
What is Magistral Medium's context window?
Magistral Medium has a 128K-token context window and a 33K-token max output — the 47th-largest context of the 49 current models we track. Source: https://docs.mistral.ai/models/magistral-medium-1-2-25-09, verified 2026-10-01.
Does Magistral Medium support vision or audio input?
No — Magistral Medium is text-only as of 2026-10-01.
Does Magistral Medium have a reasoning or extended-thinking mode?
Yes — Magistral Medium exposes a dedicated reasoning mode for multi-step problems.
When was Magistral Medium released, and what is its knowledge cutoff?
Magistral Medium was released 2026-09.
How much does Magistral Medium cost, and who provides it?
Magistral Medium is served by Mistral at $2.75/M blended tokens (3:1 input:output) — the 31st-cheapest of 49 current models. Full pricing breakdown: /llm-api-pricing/magistral-medium.