Can I run Ministral 3 3B locally?

Yes. The weights are open under Apache 2.0 and the model fits in 8 GB of VRAM in FP8; quantized further, it needs even less, making it practical on a wide range of consumer GPUs.

What does it cost and how long is the context?

Via API it is $0.10 per million tokens for both input and output. The context window is 256,000 tokens according to the model card, though the API listing shows 131,072 tokens.

How does this differ from larger Ministral 3 models?

This is the smallest model in the Ministral 3 family, optimised for edge and low-resource deployment. You trade raw capability for a model that is cheap to run and fits on modest hardware.

Ministral 3 3B 2512

5960°

#llm #mistralai #vision #agents #multilingual #local

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Mistral | 131,072 token context | $0.10/M input, $0.10/M output. The smallest model in the Ministral 3 family, Ministral 3 3B is a powerful, efficient tiny language model with vision capabilities.

Guides for this model

Ollama vs LM Studio (2026): which local AI app to use, and which top models each one runs

How much VRAM do I actually need for local AI?

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Ministral 3 3B is Mistral's smallest Ministral 3 model, a 3.4B-parameter instruct model with a built-in 0.4B vision encoder. It runs in 8 GB of VRAM in FP8, fits on edge hardware, and covers vision, function calling, and a dozen-plus languages at a price that is hard to argue with. If you need a capable tiny model you can deploy locally or hit cheaply via API, this is a serious option.

Known for

Fits in 8 GB VRAM in FP8; runs locally on modest consumer hardware
Vision encoder included: analyze images alongside text
Native function calling and JSON output for agentic pipelines
Multilingual across English, French, Spanish, German, Chinese, Japanese, Korean, Arabic and more
Apache 2.0 licence: commercial use and modification permitted

Watch out for

It is a 3B model; complex reasoning and long-form writing will hit a ceiling faster than larger siblings.
Vision works best with near-square images; very wide or thin crops degrade performance.
Mistral recommends temperature below 0.1 for production, so creative output needs deliberate tuning.

Best for

Prompts to try

Extract structured data from text

Read the following invoice text and return a JSON object with these fields: vendor_name, invoice_date, total_amount, line_items (array of description and amount). Invoice: 'Acme Supplies, 14 Nov 2025. Widget A x3 @ $12.00 each. Shipping $5.00. Total $41.00.'

Describe and classify an image

Look at the image I have attached. Identify the main subject, describe what is happening, list any text visible, and classify the image into one of these categories: product photo, document, outdoor scene, indoor scene, diagram. Keep your answer under 150 words.

Build a simple tool-calling plan

You have two tools: get_weather(city: string) and send_email(to: string, subject: string, body: string). A user says: 'Email sarah@example.com the current weather in Tokyo.' Write the sequence of tool calls you would make, in JSON, to complete this request.

Compare with

Mistral Nemo
Open-weights model
131,072 token context · $0.03/M input, $0.03/M output
Mistral Large 4
API-only model
1,048,576 token context · $0.68/M input, $2.09/M output
Mistral Small 3.2 24B Instruct 2506
Open-weights model
128,000 token context · $0.07/M input, $0.20/M output
Voxtral Small 24B 2507
Open-weights model
32,768 token context · $0.10/M input, $0.30/M output

More text models from mistralai

Questions people ask

Can I run Ministral 3 3B locally?: Yes. The weights are open under Apache 2.0 and the model fits in 8 GB of VRAM in FP8; quantized further, it needs even less, making it practical on a wide range of consumer GPUs.
What does it cost and how long is the context?: Via API it is $0.10 per million tokens for both input and output. The context window is 256,000 tokens according to the model card, though the API listing shows 131,072 tokens.
How does this differ from larger Ministral 3 models?: This is the smallest model in the Ministral 3 family, optimised for edge and low-resource deployment. You trade raw capability for a model that is cheap to run and fits on modest hardware.

Captain's verdict
A genuinely useful tiny model: vision, function calling, multilingual, Apache 2.0, and cheap enough to leave running on edge hardware without regret.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Ministral 3 3B 2512 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm mistralai vision agents multilingual local budget

Browse by Model Family