FAL AI
Connect your account, then chat with AI to run tools.
FAL AI is a machine learning platform specializing in image and video generation through AI models. Access state-of-the-art generative AI models for creating visual content, including text-to-image, image-to-image, and video generation with fast, scalable API infrastructure.
Use in your AI tool
You’ll sign in to MCPBundles when your client connects.
https://mcp.mcpbundles.com/bundle/falChat in MCPBundles Studio
This server needs a sign-in. MCPBundles handles that setup once — then chat here with saved access and a log of what ran.
Opens MCPBundles Studio with this server selected. After sign-in, chat and run tools from the same thread.
Browse all toolsLive probe refreshed Sep 14, 2026 · Endpoint host mcp.mcpbundles.com
AI SkillSKILL.md
Domain knowledge for FAL AI — workflow patterns, data models, and gotchas for your AI agent.
FAL AI
Generative image, video, audio, and utility models through fal's queue and platform catalog surfaces.
Capabilities
- Curated generators: ready-made flows for common image and video models with typed parameters.
- Live catalog: search and filter models by text and category; inspect one model's metadata, optional expanded schema for argument shapes, and list pricing.
- Cost: list unit prices and request structured cost estimates when you need batch or historical projections.
- Generic execution: run any model with a JSON argument object (blocking), or submit async jobs and poll status, fetch final JSON, or cancel.
- Movie production shortcuts: Kling v3 image-to-video (standard/pro), motion-control, Seedance 2.0, lip sync, ffmpeg merge/compose, Topaz/SeedVR upscale, ElevenLabs/MiniMax TTS, and voice clone — each pins a fal endpoint_id so agents avoid raw strings.
- Account billing: prepaid credit balance (fal equivalent of Runway org credits).
- Job wait helper: polls async queue jobs to completion after generic or curated submit flows that return request_id.
- Streaming: SSE collection for endpoints that support live output streams.
- Files: upload local or MCP resources, pull a public URL down and push bytes to fal's CDN for model inputs.
- History: list recent requests and fetch one by id from activity history.
- Docs index: search the published docs link list for topics.
Workflow
- Check billing balance before long video or batch image runs.
- Search the catalog or use a curated generator when you know the model family.
- For long renders, submit async and use the wait helper or poll status until URLs or JSON are ready.
- Chain ffmpeg merge, upscale, TTS, and lip-sync tools for multi-step production pipelines.
Gotchas
- Long video or training: prefer async submit + status/result over blocking run defaults.
- Kling clips: use Kling v3 image-to-video with tier=standard for volume 8s clips and tier=pro for hero shots; poll until complete on long renders.
- Post chain: merge clips with ffmpeg merge, then lip-sync dialogue after ElevenLabs TTS when building talking-head shots.
- Expanded model schema payloads are large — use a small limit when browsing the catalog.
- fal also ships a hosted MCP catalog with overlapping capabilities; enable both only when you need MCP-native clients alongside this bundle.
Tools discovered (40)
Showing 25 of 40 from the live probe.
Cancel Fal Endpoint
Cancel a fal queue job for endpoint_id + request_id.
Clone Fal Minimax Voice
Clone a voice from reference audio using fal-ai/minimax/voice-clone.
Compose Fal Video Layers
Composite multiple video layers with fal-ai/ffmpeg-api/compose. Pass the full layer graph as arguments_json per fal docs.
Edit Image Fal
Prompt edit a single image via fal-ai/nano-banana/edit. image_url must be a fal CDN URL — upload bytes or a public URL first if needed.
Edit Image Flux2
Higher-fidelity prompt edit via fal-ai/flux-2-pro/edit.
Edit Image Kontext
Scene-consistent prompt edit via fal-ai/flux-pro/kontext. Use for localized semantic fixes while preserving composition and lighting.
Estimate Fal Pricing
Estimate run cost via fal platform pricing API. Pass request_body_json: a JSON object per OpenAPI (e.g. {"estimate_type":"historical_api_price","endpoints":{"fal-ai/flux/dev":{"call_quantity":10}}} or unit_price variant with unit_quantity per endpoint).
Generate Fal Elevenlabs Tts
Generate speech audio with fal-ai/elevenlabs/tts/turbo-v2.5. Pair output with lip-sync for dialogue shots.
Generate Fal Image
Generate image using FAL AI FLUX model
Generate Fal Image To Video
Generate video from a source image using FAL AI Veo3 image-to-video. Pass prompt plus a public image_url. Returns request_id — poll request status until complete.
Generate Fal Kling Image To Video
Generate video from a start still using Kling v3 image-to-video. tier=standard pins fal-ai/kling-video/v3/standard/image-to-video; tier=pro pins fal-ai/kling-video/v3/pro/image-to-video. Default duration is 8 seconds for clip handoff workflows.
Generate Fal Kling Motion Control
Kling v3 motion-control on fal-ai/kling-video/v3/standard/motion-control. Use motion_prompt for dolly, pan, tilt, and subject movement grammar.
Generate Fal Minimax Tts
Generate speech with fal-ai/minimax/speech-02-hd (alternative to ElevenLabs TTS).
Generate Fal Seedance Image To Video
ByteDance Seedance 2.0 image-to-video (bytedance/seedance-2.0/image-to-video). Supports optional end_image_url for start/end frame interpolation.
Generate Fal Video
Generate video using FAL AI Veo3 model
Generate Image
Generate image using FAL AI Nano Banana model from text prompt only
Generate Image From Images
Generate new image from multiple input images using FAL AI Nano Banana model
Get Fal Account Billing
Get fal account billing summary including prepaid credit balance. Use before batch video runs to confirm sufficient credits (fal Platform API).
Get Fal Endpoint Result
Wait until the fal queue job completes (if needed) and return the final JSON result for endpoint_id + request_id.
Get Fal Endpoint Status
Poll fal queue status for a given endpoint_id and request_id from an async queue submit.
Get Fal Model Info
Get catalog metadata and unit pricing for a single fal endpoint_id (e.g. fal-ai/flux/dev). Returns description, category, model URL, and price rows. Set include_openapi_schema to true to embed fal's OpenAPI 3.0 spec for the model (request/response shapes); omit or false for a lighter response.
Get Fal Model Pricing
Return fal list pricing for a model endpoint_id (USD per billing unit).
Get Fal Request
Get detailed information about a specific FAL request
List Fal Requests
List all FAL requests from remote API
Merge Fal Videos
Merge multiple video clips in order using fal-ai/ffmpeg-api/merge-videos. Use after generating 8s Kling clips for a sequence handoff chain.
Frequently Asked Questions
What is the FAL AI MCP server?
FAL AI is a machine learning platform specializing in image and video generation through AI models. Access state-of-the-art generative AI models for creating visual content, including text-to-image, image-to-image, and video generation with fast, scalable API infrastructure. It provides 40 tools that AI agents can use through the Model Context Protocol (MCP).
How do I connect FAL AI to my AI agent?
Add the MCPBundles server URL to your MCP client configuration (Claude Desktop, Cursor, VS Code, etc.). The URL format is: https://mcp.mcpbundles.com/bundle/fal. Authentication is handled automatically.
How many tools does FAL AI provide?
FAL AI provides 40 tools that can be called by AI agents, along with a SKILL.md that gives your AI agent domain knowledge about when and how to use them.
What authentication does FAL AI require?
FAL AI uses API Key. FAL AI requires credentials. Connect via MCPBundles and authentication is handled automatically.
Setup Instructions
Connect FAL AI to any MCP client in minutes
More in machine learning
Other MCP servers in this category from the directory index
This server equips AI agents with tools to build, manage, and run automated processes. It is particu...
Aura Intelligence is an AI-powered data analytics and business intelligence platform that helps orga...
Hugging Face is an AI and machine learning platform that provides access to thousands of pre-trained...
4 tools
Materialized ComfyUI node catalog from object_info plus curated API prompt templates and offline wor...
6 toolsRun local Ollama models on the user's machine — chat, completion, embeddings, and model management t...
11 toolsRun ComfyUI image and video generation workflows on the user's local machine. Queue API-format workf...
35 tools