# Pixtral Large (24.11)

> Mistral AI's 124B open-weights vision model built on Mistral Large 2, with a 128K context window.

- Maker: Mistral AI
- Line: Pixtral
- Canonical page: https://ai-tldr.dev/models/pixtral-large/
- Updated: 2026-06-18

## In short

Pixtral Large (24.11) is Mistral AI's frontier-class multimodal model, released on 18 November 2024. It is the second model in the Pixtral line (after Pixtral 12B) and pairs a 1-billion-parameter vision encoder with the 123-billion-parameter Mistral Large 2 decoder, for roughly 124B parameters in total. It has a 128K-token context window. You can use it through Mistral AI API.

## Facts

| Fact | Value | Source |
| --- | --- | --- |
| Context window | 128K | [1] |
| Licence | Mistral Research License (MRL) for research; Mistral Commercial License for production · open weights | [1] |
| Released | 18 November 2024 | [2] |
| Parameters | 124B total (123B multimodal decoder + 1B vision encoder) | [1] |

## Sources

1. [Pixtral-Large-Instruct-2411 — Hugging Face model card](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411)
2. [Pixtral Large — official announcement](https://mistral.ai/news/pixtral-large/)

## Overview

Pixtral Large (24.11) is Mistral AI's frontier-class multimodal model, released on 18 November 2024. It is the second model in the Pixtral line (after Pixtral 12B) and pairs a 1-billion-parameter vision encoder with the 123-billion-parameter Mistral Large 2 decoder, for roughly 124B parameters in total. Mistral positioned it as a model that adds frontier-level image, document, and chart understanding on top of Mistral Large 2's text capabilities.

The model reads both text and images and ships with a 128K-token context window — enough, per Mistral, to fit a minimum of 30 high-resolution images alongside text in a single request. That makes it suited to multi-page document understanding, chart and diagram reading, and OCR-style tasks, not just single-image captioning. It is distributed with open weights: the instruction-tuned checkpoint (mistralai/Pixtral-Large-Instruct-2411) is downloadable from Hugging Face under the Mistral Research License, with a separate Mistral Commercial License for production use.

Pixtral Large was served via Mistral's API as pixtral-large-2411 (alias pixtral-large-latest) and in le Chat. Mistral has since deprecated it: its documentation lists a deprecation date of 27 February 2026 and points new projects to Mistral Medium 3.5 as the recommended replacement. The open weights remain available for research and self-hosting (Mistral recommends vLLM for inference).

## Benchmarks

- MathVista (CoT): 69.4 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- MMMU (CoT): 64 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- ChartQA (CoT): 88.1 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- DocVQA (ANLS): 93.3 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- VQAv2 (VQA Match): 80.9 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- AI2D (BBox): 93.8 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))
- MM MT-Bench: 7.4 ([source](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411))

## Strengths

- Strong document, chart, and diagram understanding: 93.3 on DocVQA (ANLS) and 88.1 on ChartQA (CoT) per Mistral's model card
- Leading visual mathematical reasoning for its generation — 69.4 on MathVista (CoT)
- Open weights under the Mistral Research License, so it can be self-hosted and fine-tuned (vLLM-supported)
- Large 128K context that accommodates roughly 30 high-resolution images plus text in one request
- Built on Mistral Large 2, so it retains that model's text-only quality while adding vision

## Best for

- Multi-page document and PDF understanding, including OCR-style extraction
- Reading and reasoning over charts, tables, and diagrams
- Visual question answering over natural images and screenshots
- Math and STEM problems presented as images (figures, plotted data)
- Self-hosted multimodal deployments where downloadable open weights are required

## How to access

- [Mistral AI API](https://docs.mistral.ai/models/model-cards/pixtral-large-24-11) — model id `pixtral-large-2411`

## Pixtral versions

- Pixtral Large (24.11) (18 November 2024) — this page
- [Pixtral 12B (24.09)](https://ai-tldr.dev/models/pixtral-12b/) (17 September 2024)

## FAQ

### Is Pixtral Large open weights?

Yes. The instruction-tuned weights (mistralai/Pixtral-Large-Instruct-2411) are downloadable from Hugging Face. They are released under the Mistral Research License (MRL) for research and educational use, with a separate Mistral Commercial License required for production commercial use.

### How big is Pixtral Large and what is it built on?

It has about 124B parameters in total: a 123B-parameter multimodal decoder based on Mistral Large 2, plus a 1B-parameter vision encoder. It supports text and image input with a 128K-token context window.

### Is Pixtral Large still available?

Mistral has deprecated it. The documentation lists a deprecation date of 27 February 2026 and recommends Mistral Medium 3.5 as the replacement. The open weights remain available on Hugging Face for research and self-hosting, with vLLM recommended for inference.

### How did Pixtral Large perform on benchmarks?

On Mistral's reported figures it scored 69.4 on MathVista (CoT), 93.3 on DocVQA (ANLS), 88.1 on ChartQA (CoT), 64.0 on MMMU (CoT), and 93.8 on AI2D. Mistral said it was the best open-weights model on the LMSys vision leaderboard at launch.

## Links

- [Pixtral Large — official announcement](https://mistral.ai/news/pixtral-large/)
- [Pixtral-Large-Instruct-2411 — Hugging Face model card](https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411)
- [Pixtral Large 24.11 — Mistral docs model card](https://docs.mistral.ai/models/model-cards/pixtral-large-24-11)
- [Mistral AI — official site](https://mistral.ai/)
- [Pixtral Large (124B) beats Llama 3.2 90B — AINews writeup](https://news.smol.ai/issues/24-11-18-ainews-pixtral-large-124b-beats-llama-32-90b-with-updated-mistral-large-2411)

Source: AI/TLDR LLM registry, https://ai-tldr.dev/models/pixtral-large/