Skip to content

Mistral AI · Chat

Pixtral Large

Mistral AI 🇫🇷ChatPixtral familyOne of its sources says the weights aren't publicly available.
From Aiml API's listing:
Pixtral Large 2502 is the February 2025 version of Mistral AI’s large multimodal model, following Pixtral Large 2411. It accepts text and image input, including documents, charts and natural images, and answers in text, with a 128K-token context window.
llminfo IQ
114.2
Maker's price
$3.00 blended / 1M
Max context
131K
Listings
25

Evidence and sources

llminfo IQ 114.2, ranked among chat models. From 15 benchmarks. Plenty of data behind it.How llminfo IQ is computedPoor value (44.0). Costs more per task than its benchmark results predict. From 1 benchmark with measured task costs. Little data behind it, so treat it as a rough estimate.How Value for money worksReleased 18 Nov 2024Maker HQ: FR (according to Mistral AI's own website)
All rankings: #453 on OpenRouter today #521 on OpenRouter this week #616 on OpenRouter this monthSources: LMArena data, Models.dev data, AI/ML API data, llminfo researchPrimary data from LMArena from Models.dev: weights availability from AI/ML API: primary mode, description from llminfo research: release date
Tags from its sources:instructvisiontoolsdocumentproprietarymultimodallong-context

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 04:34 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 9 maker's prices; 1 market median price.

Input & output

as stated by the model's providers
Text
Images
Files
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured output

Listings

 Mistral AIpixtral-largeSource: GenAI Prices dataNot published by the provider$2.00$6.00——
 Cortecs AIpixtral-large-2502 · 2502Sources: Cortecs EU Router data, Models.dev data128K——
 Amazon Web Servicesmistral.pixtral-large-2502-v1:0 · 2502Sources: AWS Bedrock Runtime data, GenAI Prices data, LiteLLM data, Models.dev data128K$2.00$6.00——
 Eden AIInference by Amazon Web ServicesRegions: 🇺🇸 usamazon/mistral.pixtral-large-2502-v1:0@us · 2502Source: Models.dev data128K$2.00$6.00——
 Eurouterpixtral-large-2502 · 2502Source: EURouter data128K$2.00$6.00——
 Flowdotmistralai/pixtral-large-2411 · 2411Source: FlowDot data131K$2.00$6.00——
 Kilomistralai/pixtral-large-2502 · 2502Sources: Kilo AI Gateway data, Models.dev data128K$2.00$6.00——
 LlmgatewayInference by Amazon Web Servicesmistral.pixtral-large-2502-v1:0 · 2502Source: LLMGateway dataNot published by the provider$2.00$6.00——
 OpenRouterInference by Amazon Web Servicesmistralai/pixtral-large-2502 · 2502Source: OpenRouter data docs128K$2.00$6.0068422ms
 Vercelmistral/pixtral-largeSource: Helicone Costs dataNot published by the provider$2.00$6.00——
Show all 21 listings

Showing 10 of 21: the maker's own listing first, then the cheapest per 1M input tokens from each provider.

Also sold by 302ai, Llmgateway Providers, Merge Gateway, Vercel AI Gateway.

4 listings without a price are hidden.

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Vision

AI2DSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
ChartQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
DocVQASource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
MMMUSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
VQAv2Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Human preference

Reasoning

General

Math

MathVistaSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim

Data credits: Data by LLM Stats · LMArena, CC BY 4.0 (licence) · Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · SnakeBench (Greg Kamradt / ARC Prize Foundation), MIT (licence)

Around the web

  • mistralai/Pixtral-12B-2409 at main - Hugging Face
    mistralai / Pixtral-12B-2409 like 692 Follow Mistral AI_ 18.7k 9 languages vllm mistral-common License: apache-2.0 …mistralai / Pixtral-12B-2409 like 692 Follow Mistral AI_ 18.7k 9 languages vllm mistral-common License: apache-2.0 Model card Files Files and versions xet Community 43 Deploy Copy to bucket new main …

    huggingface.co

  • [Deprecated] Pixtral 12B | Mistral AI
    La Plateforme or on Le Chat Pixtral is trained to understand both natural images and …La Plateforme or on Le Chat Pixtral is trained to understand both natural images and documents, achieving 52.5% on the MMMU reasoning benchmark, surpassing a number of larger models. Section: downloa…

    mistral.ai

  • Pixtral Large - Mistral AI | Mistral Docs
    Docs & API Search docs ⌘K Getting Started Models Products Developers Admin API Search docs …Docs & API Search docs ⌘K Getting Started Models Products Developers Admin API Search docs ⌘K Toggle theme Reach out Try Studio 1. 3. Models 5. Model Cards 7. Pixtral Large Try in playground ↗ Pixtra…

    docs.mistral.ai

  • [Deprecated] Pixtral 12B | Mistral AI
    Recherche Announcing Pixtral 12B September 17, 2024 By Mistral AI team Retour au blog 8 …Recherche Announcing Pixtral 12B September 17, 2024 By Mistral AI team Retour au blog 8 min de lecture Partager cet article Heads up: this model is deprecated Pixtral 12B is no longer maintained and …

    mistral.ai

  • mistralai/Pixtral-12B-2409 · Hugging Face
    Vision Encoder Supports variable image sizes Leading performance in its weight class on multimodal tasks …Vision Encoder Supports variable image sizes Leading performance in its weight class on multimodal tasks Maintains state-of-the-art performance on text-only benchmarks Sequence length: 128k License: …

    huggingface.co

  • README.md · mistralai/Pixtral-12B-Base-2409 at 845d26d58ae88e30a2d3e5c172baa7916338e009
    We recommend using Pixtral with the vLLM library to implement production-ready inference pipelines with Pixtral. …We recommend using Pixtral with the vLLM library to implement production-ready inference pipelines with Pixtral. ** Installation ** Make sure you install vLLM >= v1.6.2 : pip install --upgrade vllm A…

    huggingface.co

  • GitHub - mistralai/mistral-evals · GitHub
    Host a model using vLLM To install vLLM, follow the directions here . vllm serve …Host a model using vLLM To install vLLM, follow the directions here . vllm serve mistralai/Pixtral-12B-2409 --config_format mistral --tokenizer_mode "mistral" Step 2 : Evaluate hosted model. python -…

    github.com

  • README.md · mistralai/Pixtral-Large-Instruct-2411 at main
    mistral_common >= 1.5.0 installed: pip install --upgrade mistral_common You can also make use of a …mistral_common >= 1.5.0 installed: pip install --upgrade mistral_common You can also make use of a ready-to-go docker image or on the docker hub . Server (Image) We recommend to use Pixtral-Large-Ins…

    huggingface.co

  • mistralai/Pixtral-Large-Instruct-2411 · Hugging Face
    Pixtral-Large-Instruct-2411 is a 124B multimodal model built on top of Mistral Large 2, i.e., Mistral-Large-Instruct-2407 …Pixtral-Large-Instruct-2411 is a 124B multimodal model built on top of Mistral Large 2, i.e., Mistral-Large-Instruct-2407 . Pixtral Large is the second model in our multimodal family and demonstrates…

    huggingface.co

Also known as: amazon/mistral.pixtral-large-2502-v1:0, amazon/mistral.pixtral-large-2502-v1:0@us, aws-bedrock/pixtral-large-2502, eu.mistral.pixtral-large-2502-v1:0, mistral.pixtral-large-2502-v1:0, mistral/pixtral-large, mistral/pixtral-large-latest, mistralai/pixtral-large-2411, mistralai/pixtral-large-2502, pixtral-large-2411, pixtral-large-2502, us.mistral.pixtral-large-2502-v1:0, vercel_ai_gateway/mistral/pixtral-large

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works