Can I run Mistral Small 4 locally?

Yes — the weights are open under Apache 2.0 and available on Hugging Face at mistralai/Mistral-Small-4-119B-2603. Mistral also publishes an NVFP4 quantised checkpoint and an Eagle speculative-decoding head to cut memory and improve speed, though 119B total parameters still demands substantial hardware.

What does Mistral Small 4 cost via API?

Mistral prices it at $0.15 per million input tokens and $0.60 per million output tokens, with a 256k token context window.

How is Mistral Small 4 different from earlier Mistral Small models?

It merges three previously separate model lines — the instruct, reasoning (Magistral), and coding (Devstral) families — into one model with a switchable reasoning mode, vision input, and agentic function calling, while running significantly faster than Mistral Small 3.

Mistral Small 4

3269°

#llm #mistralai #vision #reasoning #coding #agents

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 | 262,144 token context | $0.15/M input, $0.60/M output. Mistral Small 4 is the next major release in the Mistral Small family, unifying the capabilities of several flagship Mistral models into a single system. It combines strong reasoning from Magistral, multimodal understanding from Pixtral, and agentic coding capabilities from Devstral, enabling one model to handle complex analysis, software development, and visual tasks within the same workflow.

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

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

Mistral Small 4 is a 119B mixture-of-experts model from Mistral AI that rolls reasoning, vision, and agentic coding into one system under an Apache 2.0 licence. With only 6.5B parameters active per token, it runs faster than its size suggests, and the 256k context window handles long documents without fuss. If you want a single open-weights model that can switch between quick chat and deliberate reasoning on demand, this is a serious candidate.

Known for

Per-request reasoning toggle: fast mode or deliberate reasoning, your call.
Vision input built in — attach images for analysis or document extraction.
Native function calling and JSON output for agentic and tool-use workflows.
Apache 2.0 licence: commercial use, fine-tuning, self-hosting all permitted.
40% faster end-to-end completions and 3× the throughput of Mistral Small 3.

Watch out for

119B total parameters means local deployment needs serious hardware; check VRAM before committing.
Output is text only — no image generation or audio output despite multimodal input.
Reasoning mode is only worth enabling for complex prompts; leaving it on by default adds latency.

Best for

Prompts to try

Debug a failing agent loop

I have a Python agent that calls a search tool, then a summarisation tool, then writes a report. After the first tool call it stops and returns an empty string. Here is the code: [paste yours]. Walk me through what is likely going wrong and rewrite the loop so it completes all three steps reliably.

Reason through a tricky plan

A small software team has three engineers. Engineer A can only review code, not write it. Engineer B writes fast but introduces bugs that A catches. Engineer C writes slowly but cleanly. A critical feature is due in five days. Reason step by step through how to assign work to ship on time with the fewest defects.

Extract data from an image

I am attaching a screenshot of an invoice. Extract the vendor name, invoice number, line items with quantities and unit prices, subtotal, tax amount, and total due. Return the result as a JSON object with those exact keys.

Compare with

Mistral Large 4
API-only model
1,048,576 token context · $0.68/M input, $2.09/M output
Mistral Large
API-only model
128,000 token context · $2.00/M input, $6.00/M output
Mistral Medium 3.5
API-only model
262,144 token context · $1.50/M input, $7.50/M output
Mistral Medium 3
API-only model
131,072 token context · $0.40/M input, $2.00/M output

More text models from mistralai

Questions people ask

Can I run Mistral Small 4 locally?: Yes — the weights are open under Apache 2.0 and available on Hugging Face at mistralai/Mistral-Small-4-119B-2603. Mistral also publishes an NVFP4 quantised checkpoint and an Eagle speculative-decoding head to cut memory and improve speed, though 119B total parameters still demands substantial hardware.
What does Mistral Small 4 cost via API?: Mistral prices it at $0.15 per million input tokens and $0.60 per million output tokens, with a 256k token context window.
How is Mistral Small 4 different from earlier Mistral Small models?: It merges three previously separate model lines — the instruct, reasoning (Magistral), and coding (Devstral) families — into one model with a switchable reasoning mode, vision input, and agentic function calling, while running significantly faster than Mistral Small 3.

Captain's verdict
A rare open-weights generalist that earns its keep across coding, vision, and reasoning without forcing you to juggle three separate models.

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.

TEXT / LLM

Mistral Nemo

TEXT / LLM

GPT 3.5 Turbo (older v0613)

TEXT / LLM

GLM Flash Latest

TEXT / LLM

AI Models for Image, Video & Text Generation

Mistral Small 4 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 reasoning coding agents long context

Browse by Model Family