Magistral Medium
At a glance
- Price per 1M tokens (input)
- $1.50
- Price per 1M tokens (output)
- $7.50
- Context
- 262,144 tokens
- Free access
- yes (see below)
Refreshed daily; data verified September 29, 2026. Published August 12, 2026, last price change September 4, 2026.
Magistral Medium is a multimodal model from Mistral AI designed for developers and enterprises that need to process visual information alongside text, while also automating workflows through structured actions. It excels at tasks like document analysis, image-based question answering, and building agents that can call external APIs or tools. The model's step-by-step reasoning capability makes it suitable for complex problem-solving where transparency and logical progression are valued. Mistral AI's approach emphasizes efficiency and controllability, offering a model that balances capability with operational practicality for production environments.
Specifications & pricing
| Input (per 1M tokens) | $1.50 |
|---|---|
| Output (per 1M tokens) | $7.50 |
| Cache read (per 1M tokens) | $0.15 |
| Context window | 262,144 tokens |
| Max output | 262,144 tokens |
| Capabilities | images, tool calling, reasoning |
LiteLLM community dataset (MIT), verified September 29, 2026. Official Mistral AI pricing.
What Magistral Medium would cost on your workload — run it through the cost calculator →
Price history
| Date | Input / 1M | Output / 1M |
|---|---|---|
| August 9, 2026 (first record) | $2.00 | $5.00 |
| September 4, 2026 | $1.50 | $7.50 |
Where to try Magistral Medium for free
- Mistral AI offers a free chat — Le Chat (free plan). A vendor's free chat may run a different model from the same family — the exact model is not guaranteed.
Frequently asked questions
What practical business problems does Magistral Medium solve?+
It is well-suited for automating document review, extracting insights from scanned images or diagrams, and building assistants that can take actions like querying databases or updating records through tool calling. Its step-by-step reasoning helps in audit trails and debugging, making it useful for regulated industries or complex decision support.
How do I start integrating Magistral Medium into my existing systems?+
You can access it via Mistral AI's API or through major cloud platforms that host the model. The API supports standard REST calls for text and image inputs, and you can define custom functions for tool calling. Start with a small pilot project, such as a document classification workflow, and use the built-in reasoning traces to validate outputs.
How does this model differ from other models in Mistral's lineup?+
Unlike pure text models, Magistral Medium adds native image understanding, allowing it to process photos, charts, and handwritten notes. Compared to larger flagship models, it is designed for lower latency and cost efficiency, making it a pragmatic choice for high-volume tasks that still require multimodal reasoning and tool use. It sits between lightweight text-only models and the most powerful frontier models.
What are the main limitations I should be aware of?+
While it handles many visual tasks, it may not match the fine-grained detail of specialized vision models on very complex or cluttered images. Tool calling requires you to define clear function schemas and handle edge cases in responses. Step-by-step reasoning increases output length, which can affect response time, so you should test latency for your specific use case.
Compare with others
Org chart: how to move your company onto AI
A practical map: which company roles and processes AI agents can take over, where to start, and in what order to roll it out.