Meta Llama 4 Scout 17B Instruct

Meta Llama 4 Scout 17B Instruct is a AWS Bedrock model available in Agent Serve. Meta Llama 4 Scout 17B Instruct is a AWS Bedrock model tracked in Agent Serve. It supports a 10M token context window. Pricing starts at $0.17/1M input tokens and $0.66/1M output tokens. Key capabilities include Temperature 0-1, Tool choice, Structured outputs.

Meta Llama 4 Scout 17B Instruct starts at $0.17/1M input tokens, and $0.66/1M output tokens.

Meta Llama 4 Scout 17B Instruct supports a context window of 10M tokens in Agent Serve. In an agent, this determines how much conversation history, tool outputs, and retrieved documents the model can hold in a single call.

Meta Llama 4 Scout 17B Instruct supports the following capabilities in Agent Serve: Temperature 0-1, Tool choice, Structured outputs, Max output 8k.

Best for long-context retrieval, large documents, and high-memory workflows. When used in an Agent Serve workflow, it can be selected in any Agent block from the model picker.