Models·Meta

Llama 4 Scout: What It Does and What It Costs

meta-llama/llama-4-scout

Meta’s open-weight model for text and image input, with a large context window.

$0.10Input, per million tokens
$0.30Output, per million tokens
1.31MContext window
16KLongest answer

1.31M tokens of context — about 983,040 words, on a scale that ends at 2M.

What it can do

  • Vision
  • Tool use
  • Open weights
  • Takes text
  • Takes image
  • Gives text

Released 5 Apr 2025 · checked 3 Oct 2026.

About it

Llama 4 Scout is a Meta model that accepts text and images and produces text. It has open weights, tool support, and a context window of 1,310,720 tokens. Listed token prices are $0.1 per million input tokens and $0.3 per million output tokens.

Review

In a line

Llama 4 Scout is a Meta language model that takes text or image input and returns text. Its weights are open.

What it's good at

It suits tasks that involve text, images, or using tools. It can take images as input, and tool support lets it work with tools.

Its 1,310,720-token context window lets it take in a large amount of material at once. It is one of the vision language models and is listed among open-weight models.

Price in plain terms

The listed price is $0.1 per million input tokens and $0.3 per million output tokens. A token is a small piece of text the model processes or produces.

Reading a page of about 500 words costs under 0.01 cents. A chat turn with about 1,000 input tokens and 500 output tokens costs about 0.025 cents; 1,000 such turns cost about $0.25.

Context and limits

Scout can take up to 1,310,720 tokens in a conversation or task. That is the amount it can consider at once. It can produce up to 16,384 tokens in one answer.

It returns text, even when the input includes an image.

Ways to use it

Its weights are open. The listed model identifier is meta-llama/llama-4-scout.

Verdict

Scout is worth considering if you need a model that can take both text and images, use tools, or handle a large amount of input at once. The listed prices come with example costs for sample workloads.

Look elsewhere if you need image output, since Scout returns text, or if you need details about specific tools before choosing.

Verdict

Scout is an option to compare if your work includes text and image inputs, tool use, or very long material. The listed costs include examples for small workloads, and the model has open weights. It may not suit you if you need the model to create images or need a named tool before deciding.

  • Accepts both text and image input.
  • Open weights are available.
  • Context window holds up to 1,310,720 tokens.
  • Supports tools.
  • Listed prices are $0.1 in and $0.3 out per million tokens.
  • Returns text only, not image output.
  • Maximum answer length is 16,384 tokens.
  • Specific supported tools are not named.
  • Pick it ifPeople working with text and image inputs; Tasks that need a very large context window; Users seeking an open-weight model with tool support
  • Skip it ifYou need the model to generate images; You require a specific named tool

Pricing

Listed token pricing is $0.1 per million input tokens and $0.3 per million output tokens. A page costs under 0.01 cents; 1,000 sample chat turns cost about $0.25.

Questions

Who makes Llama 4 Scout?

Meta makes Llama 4 Scout. It is part of the meta-llama family.

Can Scout understand images?

It accepts text and images as input, and its output is text.

How much can it read at once?

Its context window is 1,310,720 tokens, the amount it can consider at once.

How long can one answer be?

It can produce up to 16,384 tokens in one answer.

What does Scout cost?

The listed price is $0.1 per million input tokens and $0.3 per million output tokens. About 1,000 sample chat turns cost $0.25.

Can Scout use tools?

Yes, tool support is listed, but the specific tools are not named.

On these boards

More from Meta