Project AURIONAI Use-Cases Research & Innovation
Mistral AI
multimodal LLM

Pixtral Large

Mistral AI 124B multimodal model that understands documents, charts and natural images while keeping Mistral Large 2 text performance.

Mistral AIMistral AIFrancePixtralFamily
Benchmark summary from the Pixtral Large announcement.
Levelmodel
Classmultimodal LLM
Parameters124B
Context128,000
Released18 Nov 2024
Opennessopen weights restricted
Runs onmulti-GPU node
Statusactive

Benchmark Results

vendor-reported
AI2Dmultimodal
BBoxvendor-reported
93.8
050100

Score in %. Higher is better.

Pixtral Large model card18 Nov 2024 huggingface.co
MMMUmultimodal
CoTvendor-reported
64
050100

Score in %. Higher is better.

Pixtral Large model card18 Nov 2024 huggingface.co
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).

Lineage

continued pretrain
Mistral-Large-Instruct-2407 (Mistral Large 2) to Pixtral Large: continued pretrain, medium confidenceMistral-Large-Instruct-2407 (Mistral Large 2)Mistral-Large-Instruct-240…Mistral AIPixtral LargePixtral LargeMistral AI, Nov 2024
Mistral AIMistral-Large-Instruct-2407 (Mistral Large 2) is the parent (continued pretrain)medium confidencemodel card
Where this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 2 models. Confidence and evidence are the notes' own.

R&D Topics

Document, OCR and chart understandingexemplar

Runtimes

vLLM

Identity

Developer
Mistral AI
Family
Pixtral
Released
18 Nov 2024 sourcesannouncement page
Weights published
18 Nov 2024
Record checked
11 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
not in the notes 123B multimodal decoder plus 1B vision encoder, built on Mistral Large 2
Total parameters
124B sources
Active parameters
–
Checkpoint parameters
–
Context length
128,000 tokens
Input
text, image
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights restricted
Licence
Mistral AI Research License (MRL-0.1) sources
Access routes
API, open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
multi-GPU node
Footprint
vLLM serving example uses tensor-parallel-size 8; commercial use needs a separate Mistral commercial licence
Runtimes
vLLM
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
Hugging Face language tags: en, fr, de, es, it, pt, zh, ja, ru, ko
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Industry Use Cases

No industry use case links this model.

Sources

  1. mistral.ai/news/pixtral-large https://mistral.ai/news/pixtral-large/
  2. huggingface.co/mistralai/Pixtral-Large-Instruct-2411 https://huggingface.co/mistralai/Pixtral-Large-Instruct-2411
  3. mistral.ai/licenses/MRL-0.1.md https://mistral.ai/licenses/MRL-0.1.md

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Pixtral Large", https://aurion.semai.id/models/mistral-ai-pixtral-large/, records of October 2026.