# Magistral 1.2 (Small & Medium)

> Mistral's reasoning models gain vision and a benchmark bump

- Maker: Mistral AI
- Line: Magistral
- Canonical page: https://ai-tldr.dev/models/magistral-1-2/
- Updated: 2026-06-18

## In short

Magistral 1.2 (Small & Medium) is Mistral's reasoning models gain vision and a benchmark bump. Mistral AI released it on 17 September 2025 as an open-weight model. It has a 128K-token context window. You can use it through Mistral AI (La Plateforme).

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 128K (max output 128K (shared context budget; no separate output cap published)) | [1] |
| Input price | $2.00 / 1M tokens (Magistral Medium); $0.50 / 1M tokens (Magistral Small) | [2] |
| Output price | $5.00 / 1M tokens (Magistral Medium); $1.50 / 1M tokens (Magistral Small) | [2] |
| Licence | Apache 2.0 (Magistral Small 1.2); proprietary (Magistral Medium 1.2) · open weights | [1] |
| Released | 17 September 2025 | [3] |
| Parameters | 24B (Magistral Small 1.2); undisclosed (Magistral Medium 1.2) | [1] |

## Sources

1. [Magistral Small 1.2 — Hugging Face model card](https://huggingface.co/mistralai/Magistral-Small-2509)
2. [Magistral 1.2 (Small & Medium) pricing](https://mistral.ai/pricing)
3. [Magistral — Mistral AI (line announcement)](https://mistral.ai/news/magistral/)

## Overview

Magistral 1.2 is Mistral AI's reasoning line, shipped on 17 September 2025 as a pair of models: the open-weight Magistral Small 1.2 (a 24-billion-parameter model under Apache 2.0, distributed as mistralai/Magistral-Small-2509) and the closed, API-only Magistral Medium 1.2 (magistral-medium-2509). Both are 'thinking' models that work through problems with an explicit chain-of-thought trace before answering, positioning them as Mistral's answer to OpenAI's o-series and DeepSeek-R1.

The headline change in 1.2 is multimodality: both models now carry a vision encoder, so they reason over images as well as text — useful for chart reading, diagram interpretation, document layout analysis, and visual question answering. Mistral also reports roughly 15% gains on math and coding benchmarks (AIME 24/25 and LiveCodeBench) over the 1.1 release, plus better tool use (web search, code interpreter, image generation) and clearer, better-formatted responses. Both share a 128K-token context window and span 24 languages.

Magistral Small 1.2 is small enough to run locally — Mistral positions it for a single RTX 4090 GPU or a 32GB-RAM MacBook once quantized — while Magistral Medium 1.2 is the frontier-class variant that powers reasoning in Le Chat and is served through Mistral's La Plateforme API. Magistral Medium 1.2 is the more capable of the two on every published benchmark.

## Benchmarks

- AIME 2024 (pass@1) — Magistral Medium 1.2: 91.82% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- AIME 2025 (pass@1) — Magistral Medium 1.2: 83.48% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- GPQA Diamond — Magistral Medium 1.2: 76.26% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- LiveCodeBench v5 — Magistral Medium 1.2: 75% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- AIME 2024 (pass@1) — Magistral Small 1.2: 86.14% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- AIME 2025 (pass@1) — Magistral Small 1.2: 77.34% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- GPQA Diamond — Magistral Small 1.2: 70.07% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))
- LiveCodeBench v5 — Magistral Small 1.2: 70.88% ([source](https://huggingface.co/mistralai/Magistral-Small-2509))

## Pricing

- Input: $2.00 / 1M tokens (Magistral Medium); $0.50 / 1M tokens (Magistral Small) per 1M tokens
- Output: $5.00 / 1M tokens (Magistral Medium); $1.50 / 1M tokens (Magistral Small) per 1M tokens

La Plateforme API pricing. Magistral Small is also free to self-host under Apache 2.0.

Source: https://mistral.ai/pricing

## Strengths

- Transparent chain-of-thought reasoning with explicit [THINK] traces, strong on math (AIME) and competitive coding (LiveCodeBench)
- Open-weight 24B Small variant under Apache 2.0 — free to self-host, fine-tune, and run on a single consumer GPU or a 32GB MacBook
- New vision encoder in 1.2 lets both models reason over images, charts, and document layouts
- 128K-token context and broad multilingual coverage (24 languages)
- Cheaper API pricing than most frontier reasoning models ($0.50/$1.50 per 1M tokens for Small; $2/$5 for Medium)

## Best for

- Multi-step math, logic, and STEM problem solving where a visible reasoning trace is valuable
- Competitive-programming-style and algorithmic coding tasks
- Multimodal reasoning over diagrams, charts, screenshots, and document layouts
- On-device / private reasoning assistants using the open-weight 24B Small model
- Multilingual reasoning and agentic workflows with web search and code-interpreter tools

## How to access

- [Mistral AI (La Plateforme)](https://docs.mistral.ai/models/magistral-medium-1-2-25-09) — model id `magistral-medium-2509`
- [Mistral AI (La Plateforme)](https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09) — model id `magistral-small-2509`

## Magistral versions

- Magistral 1.2 (Small & Medium) (17 September 2025) — this page
- [Magistral 1.1 (Small & Medium)](https://ai-tldr.dev/models/magistral-1-1/) (July 2025)
- [Magistral 1.0 (Small & Medium)](https://ai-tldr.dev/models/magistral-1-0/) (10 June 2025)

## FAQ

### What's new in Magistral 1.2 versus 1.1?

The biggest change is multimodality: both Magistral Small 1.2 and Medium 1.2 add a vision encoder, so they can reason over images as well as text. Mistral also reports roughly 15% gains on math (AIME 24/25) and coding (LiveCodeBench) benchmarks, better tool use with web search and a code interpreter, and clearer, better-formatted responses.

### Is Magistral 1.2 open source?

Partly. Magistral Small 1.2 (24B parameters) is open-weight under the Apache 2.0 license and downloadable from Hugging Face as mistralai/Magistral-Small-2509, so you can self-host and fine-tune it commercially. Magistral Medium 1.2 is proprietary and available only through Mistral's La Plateforme API and Le Chat.

### Can Magistral Small 1.2 run locally?

Yes. At 24B parameters it is small enough to run on a single Nvidia RTX 4090 GPU, and Mistral says a quantized version fits in a MacBook with 32GB of RAM, making it one of the few reasoning models you can run fully on-device.

### How much does the Magistral 1.2 API cost?

On Mistral's La Plateforme, Magistral Medium 1.2 costs $2.00 per million input tokens and $5.00 per million output tokens, while Magistral Small 1.2 costs $0.50 input and $1.50 output per million tokens. Magistral Small is also free to run yourself under Apache 2.0.

## Links

- [Mistral AI — official site](https://mistral.ai/)
- [Magistral Small 1.2 — Hugging Face model card](https://huggingface.co/mistralai/Magistral-Small-2509)
- [Magistral Medium 1.2 — Mistral Docs](https://docs.mistral.ai/models/magistral-medium-1-2-25-09)
- [Magistral Small 1.2 — Mistral Docs](https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09)
- [Magistral — Mistral AI (line announcement)](https://mistral.ai/news/magistral/)
- [Magistral 1.2 announcement — Mistral AI on X](https://x.com/MistralAI/status/1968670593412190381)
- [Magistral — technical report (arXiv)](https://arxiv.org/abs/2506.10910)
- [Magistral Small 1.2 — Artificial Analysis](https://artificialanalysis.ai/models/magistral-small-2509)
- [Simon Willison — Magistral 1.2 notes](https://simonwillison.net/2025/Sep/19/magistral/)
- [Le Chat — try Magistral in Mistral's assistant](https://chat.mistral.ai/)
- [Magistral 1.2 (Small & Medium) pricing](https://mistral.ai/pricing)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/magistral-1-2/