Connect FLUX MCP through BFL OAuth for image generation, editing and supported video workflows. Check the billing organisation, reference inputs and current tools.
FLUX MCP is Black Forest Labs' hosted connection for image and video workflows in an MCP client. The current publisher guide covers generation, editing, variations, multi-reference work and browsing previous results.
Use the documented endpoint, https://mcp.bfl.ai, and complete OAuth sign-in with a BFL account. Select the intended organisation before starting a billed task. The provider handles usage and pricing through that account; this integration should not be described as a free local image model.
The repository includes client-specific connection examples. Clients that need a stdio bridge can use the documented mcp-remote route, which still connects to the hosted service and opens the OAuth flow.
Describe the subject, composition, intended output and changes you want from a supplied reference. For variations, identify the result to branch from and the elements that should stay consistent. Inspect the returned options before requesting another batch.
Image and video tools can have different inputs and account requirements. Check the current discovered tools and publisher reference rather than assuming every workflow is available in every client.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| generate_image | Generate 1–8 images in parallel. Covers text-to-image, edits, multi-reference composition, style transfer, inpainting-style edits, and outpainting — all… |
| vto | Virtual try-on — dress a person (or pet) in a garment, hat, sunglasses, shoes, bag, or any wearable from a reference image. Preserves face, hair, and pose… |
| generate_video | Generate 1–4 videos in parallel with FLUX 3 — text-to-video, image-to-video from keyframes, or continuation of an existing video. 5–20 seconds, synchronized… |
| enhance_video | Re-render a draft video at full quality — the same composition, seed, and prompt plan as the draft you picked. |
| generate_variations | Produce N more images in the same direction as a previous generation (identified by request_id). Defaults to 4, max 8. |
| get_history | List recent generations as a thumbnail grid with per-tile Variations / Edit / copy / download actions. Keyset pagination, date filters. |
| get_credits | Return your remaining BFL credit balance. |
| list_skills | List the official BFL prompting guides — image best practices, video direction, audio and dialogue, keyframes, product ads, archival looks, API integration. |
| get_skill | Fetch one guide by name. Long guides return an intro plus a section index; fetch the sections the task needs. |
{
"mcpServers": {
"FLUX": {
"url": "https://mcp.bfl.ai"
}
}
}Complete the publisher’s OAuth sign-in flow.
A BFL account and a compatible hosted MCP client. Sign in with OAuth and select the intended billing organisation.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.
Thousands of open models on tap — run image, video and audio generation through Replicate's hosted API.
SEO audit (400 checks), text readability and grammar check, PDF to images, image conversion.
MCP (Model Context Protocol) server for the Poof background removal API. Use AI assistants like Claude to remove backgrounds from images.
Generate and edit images, videos, and audio with 150+ models from 20+ vendors.
A Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.