Mistral Vision Models

On Roboflow's Vision Evals, Mistral Large 4 is Mistral's top model at 56.8% (#43 of 65). Compare all 4 Mistral vision models we track, 3 of them open-weight. Try 3 of them on your own images, free in the Roboflow Playground.

4 models · 3 open-weight · 3 free to try · Updated Oct 2026

About Mistral Vision Models

Mistral ships vision models on two tiers, and our catalog tracks 4 of them. 3 are open-weight: the weights are downloadable, so you can self-host under their own licenses. The other 1 is proprietary, reached through an API and billed by the provider. The open-weight side runs on Custom and Apache 2.0 licenses. The newest additions are Mistral Large 4 (Oct 2026) and Mistral Medium 3.1 (Aug 2025).

Between them, the Mistral models we list cover 11 distinct vision tasks. The widest coverage is captioning (4 models), chart question answering (4), and classification (4). For captioning, start with Mistral Large 4, Mistral Medium 3.1, or Mistral Small 3.1 24B.

On the open-weight tier, Mistral Large 4 is the largest at 1.05T total, 49B active parameters and Pixtral 12B the smallest at 12B. 2 of the 3 open Mistral models carry a permissive license, so commercial use is straightforward; the other 1 ship under terms worth reading before you deploy.

On the API tier, the one option is Mistral Medium 3.1 (Aug 2025), billed per token by the provider rather than run on your own hardware. Which tier to start on is a constraints question, not a quality one: reach for the open-weight side when you need offline inference, predictable per-image cost, or a checkpoint you can fine-tune on your own data, and for the API side when you want broad general reasoning without managing GPUs. 3 of the 4 Mistral vision models run live in the Roboflow Playground, so you can run the same image through several of them and compare the answers before committing to one.

Which Mistral Model Should You Use?

What each of the 4 Mistral vision models in our catalog is built for, and how you run it.

Mistral Large 4: Best for: Object Detection and OCR. Open weights: Custom license, 1.05T total, 49B active parameters. Self-host it or run it here. Runnable in the Playground.
Mistral Medium 3.1: Best for: OCR and Document Question Answering. Proprietary: no downloadable weights, billed per token through Mistral's API. Runnable in the Playground.
Mistral Small 3.1 24B: Best for: OCR and Document Question Answering. Open weights: Apache 2.0 license, 24B parameters. Self-host it or run it here. Runnable in the Playground.
Pixtral 12B: Best for: OCR and Document Question Answering. Open weights: Apache 2.0 license, 12B parameters. Self-host it or run it here.

Open-Source Mistral Models

3 models with downloadable weights you can self-host under their licenses (Custom and Apache 2.0). 2 run live in the Playground through hosted APIs, so self-hosting is optional.

Mistral Large 4

Mistral Large 4, nicknamed Le Chonk, is a natively multimodal mixture-of-experts model from Mistral that accepts interleaved text and image input and produces text output. It uses a granular MoE design with roughly 1.05 trillion total parameters and 49 billion active per token, reported as 52 billion when embeddings and output layers are counted, paired with a 1.6 billion parameter vision encoder and a context window of one million tokens. The model is trained from scratch on about 3,800 NVIDIA Grace Blackwell GPUs in Mistral's European data centers and supports more than 160 languages. It behaves as a hybrid instruct and reasoning system, with a reasoning effort setting that selects between direct answers and longer deliberation, alongside function calling and structured output for agentic workflows.Image understanding is a focus of this generation, covering documents, charts, technical drawings and natural scenes, and the model emits bounding box coordinates for visual grounding queries. Reported grounding results include 42 percent on Dense200 and 73 percent on the DIOR-RSVG remote sensing benchmark. Mistral describes agentic vision workflows in which the model zooms into gigapixel satellite imagery or engineering drawings to verify details, and reports coding results such as 62 percent on DeepSWE. Figures published at preview time are preliminary because the reinforcement learning phase is still in progress.

Mistral Small 3.1 24B

Mistral Small 3.1 24B, released on March 17, 2025, is an open-weight multimodal model from Mistral AI, distributed under the Apache-2.0 license. With around 24B parameters and a 128K token context window, it is available in both base and instruction-tuned (“Instruct”) variants. The model introduces vision support alongside text, enabling tasks like multimodal reasoning, captioning, and image-based Q&A.It is multilingual, supporting many languages, and is optimized for fast responses, function calling, structured dialogue, and long-context reasoning. Despite its size, the model can be run locally in quantized formats, fitting on machines with ~32GB RAM, making it accessible to developers outside large cloud setups. However, the output length is smaller than the 128K input window, meaning long generations may require chaining. In addition, using full vision features or the maximum context window significantly increases compute costs, and performance on highly complex reasoning or enterprise-scale tasks still trails larger proprietary frontier models.

Pixtral 12B

Pixtral-12B is a vision-language model introduced by Mistral AI in September 2024 under the Apache 2.0 license, designed to process both text and images in a unified context. With ~12 billion parameters in its decoder and an additional ~400 million in a custom-trained vision encoder, it supports long-context reasoning up to 128k tokens and accepts multiple images per input. Its architecture is optimized for handling variable image sizes and aspect ratios, making it flexible for diverse multimodal tasks.As Mistral’s first VLM, Pixtral-12B delivers strong performance not only on image-text reasoning benchmarks but also in text-only applications, positioning it as a versatile alternative to models like GPT-4V and LLaVA. Its open availability via Hugging Face and major cloud providers such as Amazon Bedrock and SageMaker makes it accessible for research and production. Typical use cases include document analysis, visual QA, data extraction, and multimodal assistants requiring both textual and visual understanding.

Mistral Models via API

1 proprietary model where the weights aren't downloadable: access is through each provider's API and billed by them. Try all of them free in the Playground.

Mistral Medium 3.1

Mistral Medium 3.1, released in August 2025 as the mistral-medium-2508 update, is a proprietary frontier model from Mistral AI positioned between smaller open models and high-end closed LLMs. It is multimodal, handling both text and image inputs, with a context window of ~128K tokens. Compared to Mistral Medium 3.0, the 3.1 release introduces improvements in reasoning, coding, STEM, and enterprise workflows, along with better tone control for conversational and business applications. It is designed for scalable enterprise deployments, including hybrid cloud and on-premises VPC setups. As part of Mistral’s Premier line, Medium 3.1 is a commercial-only offering: while it delivers strong accuracy and performance, trade-offs include higher costs than open-weight models, restricted fine-tuning access, and increased latency/cost for very large contexts.

Frequently Asked Questions About Mistral Vision Models

Are Mistral vision models open source?

Partly. 3 of the 4 Mistral vision models we track are open-weight (for example Mistral Large 4, Mistral Small 3.1 24B, and Pixtral 12B), downloadable and self-hostable under their licenses (Custom and Apache 2.0). The other 1 are proprietary and reached through Mistral's API.

What is the best Mistral vision model?

On Roboflow's Vision Evals, Mistral Large 4 is the highest-scoring Mistral model, ranked #43 of 65 overall at 56.8%. The overall score averages the 5 benchmark tasks, so check the task leaderboards if you need one capability.

How many Mistral vision models are on Roboflow Playground?

We track 4 live Mistral vision models, 3 open-weight and 1 available through an API. The most recent addition is Mistral Large 4, released Oct 2026.

Can I try Mistral vision models for free?

Yes. 3 of the 4 Mistral models run live in the Roboflow Playground. Upload your own image, run several models on it at once, and compare the outputs side by side. You can try it instantly, and a free account unlocks unlimited runs.

This page lists all 4 Mistral vision models in the Roboflow Playground catalog: 3 open-weight models you can self-host and 1 proprietary model accessed through an API. They cover captioning, chart question answering, and classification, among other tasks. 3 of them run live in the Roboflow Playground on your own images. Compare licenses, parameters, prices, and release dates side by side, or open any model page for full details.