generated: '2026-06-20'
method: searched
status: published
source: https://fal.ai/docs/documentation/setting-up/mcp
docs: https://fal.ai/docs/documentation/setting-up/mcp
announcement: https://blog.fal.ai/connect-your-ai-to-1-000-models-with-the-fal-mcp-server/
server:
  name: fal-ai
  transport: http
  streamable: true
  url: https://mcp.fal.ai/mcp
  auth:
    type: bearer
    header: 'Authorization: Bearer YOUR_FAL_KEY'
    note: Uses a personal fal API key. The MCP server is free; you pay only for the model runs you trigger,
      at standard API pricing.
install:
  claude_code: "claude mcp add --transport http fal-ai \\\n  https://mcp.fal.ai/mcp \\\n  --header \"\
    Authorization: Bearer YOUR_FAL_KEY\"\n"
  clients:
  - Claude Code
  - Claude Desktop
  - Cursor
  - Windsurf
tools:
- name: search_models
  group: discovery
  description: Search fal's catalog of 1,000+ models by keyword or category.
- name: get_model_schema
  group: discovery
  description: Get the full input/output parameters for any model.
- name: get_pricing
  group: discovery
  description: Check the cost of running a model before you use it.
- name: search_docs
  group: discovery
  description: Search the fal documentation for guides, examples, and API references.
- name: run_model
  group: execution
  description: Run any model and wait for the result.
- name: submit_job
  group: execution
  description: Submit a long-running job and return immediately with a request ID.
- name: check_job
  group: execution
  description: Check job status, fetch results, or cancel a running job.
- name: upload_file
  group: utility
  description: Upload a file (local path or URL) to fal's CDN for use as model input.
- name: recommend_model
  group: utility
  description: Describe what you want to build and get model recommendations.
deployment:
  mode: remote
  endpoint: https://mcp.fal.ai/mcp
  verified: probed
  probe: gated
  checked: '2026-08-12'
  source: catalog MCP census