Replicate Flux MCP provides a bridge to the Replicate API for accessing the Flux image generation model. Built by Yaroslav Boiko, this TypeScript implementation enables AI assistants to generate images through the Model Context Protocol without requiring users to manage separate API integrations. It handles configuration through a simple API token setup and is packaged as both an npm module and a Docker container, making it ideal for applications that need to incorporate AI image generation capabilities within conversation interfaces.
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AIGenerates an image based on a text prompt using the configured image model. Parameters: prompt (string), model_id (optional string), seed (optional number), go_fast (optional boolean), megapixels (optional string), num_outputs (optional number), aspect_ratio (optional string), output_format (optional string), output_quality (optional number), num_inference_steps (optional number), disable_safety_checker (optional boolean), support_image_mcp_response_type (optional boolean).
Generates multiple images based on an array of prompts using the configured image model. Parameters: prompts (string array), model_id (optional string), seed (optional number), go_fast (optional boolean), megapixels (optional string), aspect_ratio (optional string), output_format (optional string), output_quality (optional number), num_inference_steps (optional number), disable_safety_checker (optional boolean), support_image_mcp_response_type (optional boolean).
Generates multiple variants of the same image from a single prompt using the configured image model. Parameters: prompt (string), model_id (optional string), num_variants (number), prompt_variations (optional string array), variation_mode (optional string), seed (optional number), go_fast (optional boolean), megapixels (optional string), aspect_ratio (optional string), output_format (optional string), output_quality (optional number), num_inference_steps (optional number), disable_safety_checker (optional boolean), support_image_mcp_response_type (optional boolean).
Generates SVG/vector output based on a text prompt using the configured SVG model. Parameters: prompt (string), size (optional string), style (optional string).
Retrieves a list of your recent predictions from Replicate. Parameters: limit (optional number).
Gets detailed information about a specific prediction. Parameters: predictionId (string).
Runs a whitelisted Replicate model with a raw input payload. Parameters: model_id (string), input (optional record).
Runs any model hosted on Replicate by its 'owner/name[:version]' reference. Parameters: model (string), input (record), prefer_wait (optional number), return_as (optional string).
Fetches the OpenAPI input schema and description for a Replicate model. Parameters: model (string).
