Can I run Llama 4 Scout locally?

Yes, the weights are open and available on Hugging Face under the Llama 4 Community License. Bear in mind the full model is 109B parameters, so you will need substantial hardware; quantised versions ease that requirement considerably.

What does Llama 4 Scout cost via API, and how long is its context?

Pricing is $0.18 per million input tokens and $0.59 per million output tokens. The supported context window is 1,310,720 tokens via API, while the underlying model was trained to handle up to 10 million tokens.

How does Scout differ from Llama 4 Maverick?

Both are Llama 4 MoE models, but Scout activates 17B parameters with 16 experts and is positioned as the leaner, more deployment-friendly option; Maverick is the larger sibling aimed at higher-capability tasks.

Llama 4 Scout

5292°

#llm #meta-llama #vision #multilingual #long context #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.

Meta | 1,310,720 token context | $0.18/M input, $0.59/M output. Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens. Built for high efficiency and local or commercial…Meta | 1,310,720 token context | $0.18/M input, $0.59/M output. Llama 4 Scout 17B Instruct (16E) is a mixture-of-experts (MoE) language model developed by Meta, activating 17 billion parameters out of a total of 109B. It supports native multimodal input (text and image) and multilingual output (text and code) across 12 supported languages. Designed for assistant-style interaction and visual reasoning, Scout uses 16 experts per forward pass and features a context length of 10 million tokens, with a training corpus of ~40 trillion tokens. Built for high efficiency and local or commercial deployment, Llama 4 Scout incorporates early fusion for seamless modality integration. It is instruction-tuned for use in multilingual chat, captioning, and image understanding tasks. Released under the Llama 4 Community License, it was last trained on data up to August 2024 and launched publicly on April 5, 2025.

Guides for this model

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

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

Llama 4 Scout is Meta's efficient mixture-of-experts model, activating 17 billion of its 109 billion parameters per forward pass. Its headline feature is a 10-million-token context window, the longest of any open-weights model at launch, paired with native image understanding across 12 languages. It suits developers and researchers who need multimodal capability at a price well below frontier closed models, or who want to run a capable vision-language model locally.

Known for

10-million-token context window, the longest available in open weights
Native image and text input via early fusion, no bolt-on vision adapter
MoE architecture keeps active parameters at 17B, reducing inference cost
Multilingual across 12 languages, covering chat, captioning and code
Open weights under the Llama 4 Community License, local deployment allowed

Watch out for

The Llama 4 Community License is not fully permissive; check usage limits before commercial deployment.
109B total parameters means local inference demands serious hardware despite the 17B active count.
Training data cuts off August 2024; anything after that is outside what it knows.

Best for

Prompts to try

Analyse a dense document

I'm pasting a long contract below. Read the entire thing, then list every clause that creates a financial obligation on my company, the dollar amount or formula involved, and the section number where it appears. Flag any clause where the obligation is open-ended or uncapped.

Describe and reason about an image

Look at the image I've attached. Describe what is happening in precise detail, then tell me what is unusual or inconsistent about it compared to what you would normally expect to see in this kind of scene. Give your reasoning step by step.

Multilingual customer reply

A customer has written to us in French complaining that their order arrived damaged and they want a refund. Draft a professional reply in French that apologises, confirms we will process a full refund within 5 business days, and asks them to keep the damaged item for our records. Keep the tone warm but not effusive.

Compare with

Llama 3.2 3B Instruct
Open-weights model
131,072 token context · $0.05/M input, $0.33/M output
Llama 3.1 70B Instruct
Open-weights model
131,072 token context · $0.40/M input, $0.40/M output
Llama Guard 4 12B
Open-weights model
163,840 token context · $0.18/M input, $0.18/M output
Llama 4 Scout 17B 16E Instruct
Open-weights model
327,680 token context · $0.10/M input, $0.30/M output

More text models from meta-llama

Questions people ask

Can I run Llama 4 Scout locally?: Yes, the weights are open and available on Hugging Face under the Llama 4 Community License. Bear in mind the full model is 109B parameters, so you will need substantial hardware; quantised versions ease that requirement considerably.
What does Llama 4 Scout cost via API, and how long is its context?: Pricing is $0.18 per million input tokens and $0.59 per million output tokens. The supported context window is 1,310,720 tokens via API, while the underlying model was trained to handle up to 10 million tokens.
How does Scout differ from Llama 4 Maverick?: Both are Llama 4 MoE models, but Scout activates 17B parameters with 16 experts and is positioned as the leaner, more deployment-friendly option; Maverick is the larger sibling aimed at higher-capability tasks.

Captain's verdict
A rare combination of open weights, genuine vision capability and an absurd context window at a price that makes frontier closed models look overpriced.

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

Mixtral 8x22B Instruct

TEXT / LLM

Moondream 3.1

TEXT / LLM

Trinity Large Thinking

TEXT / LLM

Claude Fable 5

TEXT / LLM

AI Models for Image, Video & Text Generation

Llama 4 Scout 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 meta-llama vision multilingual long context local budget

Browse by Model Family