Magistral Medium
Mistral AI's enterprise reasoning model, released June 2025, with strong multilingual reasoning.
Key Specifications
- Context window
- 128K tokens
- Max output
- 33K tokens
- Released
- Inputs
- Text, image
- Capabilities Show details
- Reasoning Thinks before it answers, always on or as a switchable mode. Function calling Connect to external tools, APIs, and systems. Structured output Return responses in structured formats like JSON.
Frequently Asked Questions
What is Magistral Medium good for?
Multilingual reasoning over text and images: Mistral built it as its enterprise reasoning model.
When is Magistral Medium not a good fit?
Any new work: Mistral retired it from its API in July 2026. Closed weights, so no self-hosting, and its maximum output caps long reasoning traces.
Can I self-host Magistral Medium?
No. Mistral does not publish the weights; Magistral Medium is available only through hosted APIs.
More from Mistral
| Model | Context | Input /1M | Output /1M |
|---|---|---|---|
|
|
256K | $0.20 | $0.20 |
|
|
256K | $0.10 | $0.10 |
|
|
256K | $0.15 | $0.15 |
|
|
256K | $0.50 | $1.50 |
|
|
256K | $1.50 | $7.50 |
|
|
256K | $0.15 | $0.60 |
|
|
128K | $0.30 | $0.90 |
|
|
128K | ||