GTWY logo
    Groq logoGroq

    meta-llama/llama-4-scout-17b-16e-instruct

    groq/meta-llama/llama-4-scout-17b-16e-instruct

    ChatVisionToolsSystem prompt

    Context

    —

    Max output

    8K

    Input / 1M

    $0.11

    Output / 1M

    $0.34

    Modality

    Chat

    Cutoff

    Aug 2024

    About this model

    Llama 4 Scout is Meta's natively multimodal model that enables text and image understanding. With a 17 billion parameter mixture-of-experts architecture (16 experts), this model offers industry-leading performance for multimodal tasks like natural assistant-like chat, image recognition, and coding tasks. With a 128K token context window and support for 12 languages (Arabic, English, French, German, Hindi, Indonesian, Italian, Portuguese, Spanish, Tagalog, Thai, and Vietnamese), the model delivers exceptional capabilities, especially when paired with Groq for fast inference.

    Best suited for

    • Knowledge-intensive reasoning, long-context analysis, large-scale document processing, multimodal (text/image) tasks, and deep research workflows requiring massive context windows.

    Built-in tools

    Gtwy web search

    Hosted by the gateway — enable them per request without wiring your own endpoint.

    Capabilities

    Vision

    Accepts images alongside text in the same message.

    Tools

    Native function calling, so agents can invoke your endpoints.

    System prompt

    Honours a dedicated system role, separate from the user turn.

    Supported parameters

    creativity_level

    Controls the creativity of responses. Higher values (e.g., 0.7) increase creativity; lower values (e.g., 0.2) make responses more predictable.

    max_tokensMax Tokens Limit

    Specifies the maximum number of text units (tokens) allowed in a response, limiting its length.

    log_probability

    If true, returns the log probabilities of each output token returned in the content of message.

    repetition_penalty

    The `frequency_penalty` controls how often the model repeats itself, with higher positive values reducing repetition and negative values encouraging it.

    novelty_penalty

    Discourages responses that are too similar to previous ones.

    stop

    This parameter tells the model to stop generating text when it reaches any of the specified sequences (like a word or punctuation)

    tools

    Lists tool definitions or capabilities available to the model.

    tool_choice

    Decides whether to use tools or just the model for generating responses.

    response_type

    Defines the format or type of the generated response.

    stream

    Sends the response in real-time as it's being generated.