Startups get up to $50k in free compute credits.

Image, video & 3D inference

Run cutting-edge AI models for image and video generation on Modal's serverless GPUs.

“Modal's beautifully designed devex makes writing distributed code as easy as writing scripts. We can turn research code into something that's deployed to production in hours instead of days.”

Prady Modukuru, Co-Founder & CEO

“As a startup, you need to iterate on things quickly. So it's really helpful when the developer experience and development speed is suddenly like 5x or 10x. It's a lot easier to deploy a ComfyUI workflow because Modal is serverless, so it auto-scales really well.”

Coco Mao, CEO & Cofounder

“We are constantly shipping the most cutting-edge creative AI machine learning techniques so our customers have access to the best creative models. Modal's has helped us streamline the process from idea to deployed pipeline, allowing us to both deploy quickly & scale rapidly.”

Weber Wong, Founder

One-stop shop for media generation

Deploy custom models

From Stable Diffusion to Flux to OpenSora, deploy any model, including custom ones, on Modal.

Advanced techniques

Fine-tuning with LoRA, personalization with Dreambooth, image manipulation with Controlnet - do it all on Modal!

Performant and efficient

Premium hardware

Leverage Modal's fleet of A100 and H100 GPUs to run resource-intensive models like Stable Diffusion XL or complex video processing tasks.

Lightning-fast cold starts

Achieve near-instant model initialization for responsive image generation.

Infinite scalability, zero idle costs

Automatically scale to hundreds of GPUs during traffic spikes, then scale back to zero when idle.

Ecosystem integrations

Framework support

Run popular frameworks like ComfyUI, A1111 and Fooocus.

Share prototypes instantly

Publish a public HTTP endpoint or server with a single Python decorator.

Try it out

Custom pet art from Flux with Hugging Face and Gradio

Fine-tune an image generation model on pictures of your pet.

Serve diffusion models

Serve Flux on Modal with a number of optimizations for blazingly fast inference.

Star in custom music videos

Fine-tune a Wan2.1 video model on your face and run it in parallel

Bring images to life

Prompt a generative video model to animate an image.

Render a video with Blender

Render an animated 3D scene using Blender's Python interface across many processors in parallel.

ControlNet playgrounds

Play with all 10 demo Gradio apps from the ControlNet project.

Run vision-language models with SGLang

Ask questions about images and get back answers from a multimodal model.

Real-time object detection

Create a web endpoint that leverages HuggingFace models to do object detection in real-time.

Stable Diffusion CLI, API, and Web UI

Generate images with diffusion models from the interface of your choosing.

Fine-Tuning and Inference for Computer Vision with YOLO

Customize and deploy lightning-fast object detection models.

Document OCR job queue

Use Modal as an infinitely scalable job queue that can service async tasks from a web app.

LoRAs Galore

Create a LoRA Playground with Modal, Gradio, and S3

Ship your first app in minutes.

$30 / month free compute

© Modal 2025