fal-generate

Generate images and videos using state-of-the-art AI models on fal.ai. All requests use the queue system by default for reliability: Benefits: For long video jobs, use --async to get request_id immediately: Supported file types: Upload flow (two-step): Max file size: 100MB (simple upload) Mode Options: Queue Operations: Advanced: To discover the best and latest models, use the search API: Or use the search_models MCP tool with relevant keywords. Categories: text-to-image , image-to-image , text-to-video , image-to-video , text-to-speech , speech-to-text IMPORTANT: Fetch schema to see exact parameters for any model. API Endpoint: Queue Submit Response: Final Result: Images: Videos: Async Submission: Control how long generated files remain accessible: Solution: Use --status and --result to check manually, or increase --timeout . Solution: Run ./generate.sh --add-fal-key or export FAL_KEY=your_key . Go to claude.ai/settings/capabilities and add *.fal.ai to allowed domains. - Long-running tasks (video) won't timeout - Can check status anytime - Can cancel queued requests - Results retrievable even if connection drops - Images: jpg, jpeg, png, gif, webp - Videos: mp4, mov, webm - Audio: mp3, wav, flac