fal.ai REST API
Fast, scalable inference for AI image and video models
fal.ai provides serverless GPU infrastructure for running AI models with ultra-low latency. The platform specializes in image generation, video synthesis, and real-time AI inference with support for Stable Diffusion, FLUX, and custom models. Developers use fal.ai for production-grade AI applications requiring speed, reliability, and scalability without managing infrastructure.
https://fal.run
API Endpoints
| Method | Endpoint | Description |
|---|---|---|
| POST | /fal-ai/flux/dev | Generate images using FLUX.1 [dev] model with advanced prompt following |
| POST | /fal-ai/flux-pro | Generate high-quality images using FLUX.1 [pro] model |
| POST | /fal-ai/stable-diffusion-xl | Generate images using Stable Diffusion XL with custom parameters |
| POST | /fal-ai/fast-sdxl | Ultra-fast SDXL generation optimized for speed |
| POST | /fal-ai/recraft-v3 | Generate images with Recraft V3 for design and branding |
| POST | /fal-ai/aura-flow | Generate images using AuraFlow open-source model |
| POST | /fal-ai/stable-video | Generate videos from images using Stable Video Diffusion |
| POST | /fal-ai/lora | Apply LoRA adapters to base models for custom styles |
| POST | /fal-ai/face-to-sticker | Convert face photos into cartoon stickers |
| POST | /fal-ai/remove-background | Remove backgrounds from images using AI segmentation |
| POST | /fal-ai/upscale | Upscale images to higher resolution using AI |
| GET | /fal-ai/queue/requests/{request_id} | Get status and result of a queued generation request |
| GET | /fal-ai/queue/requests/{request_id}/status | Check the current status of a request without fetching full result |
| POST | /fal-ai/image-to-image | Transform images based on text prompts using img2img |
| POST | /fal-ai/controlnet | Generate images with spatial conditioning using ControlNet |
Sponsor this page
AvailableReach developers actively building with fal.ai. See live pageview data and self-serve checkout — your slot goes live in minutes.
View inventory & pricing →Code Examples
curl -X POST https://fal.run/fal-ai/flux/dev \
-H "Authorization: Key YOUR_FAL_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "A serene mountain landscape at sunset, photorealistic",
"image_size": "landscape_4_3",
"num_inference_steps": 28,
"guidance_scale": 3.5,
"num_images": 1,
"enable_safety_checker": true
}'
Use fal.ai from Claude / Cursor / ChatGPT
Get a hosted MCP endpoint for fal.ai. Paste your fal.ai API key, copy back one URL, drop it into Claude Desktop, Cursor, or any AI client that supports remote MCP. Your AI calls fal.ai directly with your credentials — no local install, works on mobile.
generate_image
Generate images from text prompts using FLUX, SDXL, or other fal.ai models with customizable parameters
transform_image
Apply image-to-image transformations, background removal, upscaling, or style transfer to existing images
create_video
Generate short videos from images or text using Stable Video Diffusion and other video models
check_generation_status
Monitor the status and retrieve results of asynchronous AI generation requests
apply_lora_style
Apply custom LoRA adapters to base models for specific artistic styles or subjects
Connect in 60 seconds
Paste your fal.ai key → get an MCP URL → paste into Claude/Cursor. Hosted by IOX, encrypted at rest.
Connect fal.ai to your AI →