Version control
for AI prompts.

Create, version, test and reuse your prompts across multiple AI models – just like code.

import { Promptil } from "promptil";

const promptil = new Promptil({
  apiKey: process.env.PROMPTIL_API_TOKEN
  });

const output = await promptil.prompts.run({
  prompt: {
    id: "|",
    version: "latest",
    variables: {
      "|": "|"
    }
  }
  });

console.log(output);

AI Generated Image

a woman relaxing in a French bookstore

scene
500M+
Runs
10K+
Models
50K+
Developers

AI generated images

A poolside patio at sunset

black-forest-labs/flux-2-pro

A banana shaped armchair ERD

google/nano-banana-pro

Woman relaxing in a French bookstore

bytedance/seedream-4

Futuristic robot looking into distance

black-forest-labs/flux-pro

Abstract painting of a sunrise

black-forest-labs/flux-pro

Cyberpunk city skyline

google/imagen-4

A poolside patio at sunset

black-forest-labs/flux-2-pro

A banana shaped armchair ERD

google/nano-banana-pro

Woman relaxing in a French bookstore

bytedance/seedream-4

Futuristic robot looking into distance

black-forest-labs/flux-pro

Abstract painting of a sunrise

black-forest-labs/flux-pro

Cyberpunk city skyline

google/imagen-4

With Promptil you can

Thousands of models contributed by our community. Not just demos — they all actually work and have production-ready APIs.

B
black-forest-labsOfficial

flux-2-pro

High-quality image generation and editing with support for eight reference images

3.4M runs
G
googleOfficial

imagen-4

Google's flagship image generation model

6M runs
B
bytedanceOfficial

seedream-4

Text-to-image generation and precise editing at up to 4K resolution

14.1M runs
B
black-forest-labsOfficial

flux-schnell

The fastest image generation model tailored for local development

553.3M runs
O
openaiOfficial

gpt-5.1

The best model for coding and agentic tasks

44K runs
G
googleOfficial

veo-3.1

High-quality video generation with synced audio support

98.8K runs

How It Works

You can get started with any model with just one line of code. But as you do more complex things, you can fine-tune models or deploy your own custom code.

Step 01

Run Models

Our community has published thousands of models that are ready to use in production. You can run these with one line of code.

import promptil

output = promptil.run(
  "black-forest-labs/flux-dev",
  input={
    "prompt": "An astronaut riding a rainbow unicorn",
  }
)

print(output)
Step 02

Fine-tune with Your Data

You can improve models with your own data to create new models that are better suited to specific tasks.

training = promptil.trainings.create(
  destination="username/drone-art"
  version="ostris/flux-trainer:e440...",
  input={
    "steps": 1000,
    "input_images": "images.zip",
    "trigger_word": "TOK",
  },
)
Step 03

Deploy Custom Models

You aren't limited to the models on Promptil: you can deploy your own custom models using Cog.

# cog.yaml
build:
  gpu: true
  python_version: "3.10"
  python_packages:
    - "torch==2.0"

predict: "predict.py:Predictor"

Scale on Promptil

Thousands of businesses are building their AI products on Promptil. Your team can deploy an AI feature in a day and scale to millions of users.

Automatic Scaling

If you get a ton of traffic, Promptil scales up automatically to handle the demand. If you don't get any traffic, we scale down to zero.

Pay for What You Use

Promptil only bills you for how long your code is running. You don't pay for expensive GPUs when you're not using them.

Forget About Infrastructure

Deploying machine learning models at scale is hard. API servers, weird dependencies, enormous model weights, CUDA, GPUs, batching... We handle it.

Logging & Monitoring

Metrics let you keep an eye on how your models are performing, and logs let you zoom in on particular predictions to debug.

Transparent Pricing

Only pay for the resources you use. No hidden fees.

CPU$0.000100/sec
Nvidia T4 GPU$0.000225/sec
Nvidia A100 (80GB) GPU$0.001400/sec
Estimated monthly cost
$0
Calculated for 1,000 predictions/day