How to beginner · 3 min read

How to use Stable Diffusion on Replicate

Q: How to use Stable Diffusion on Replicate

Use the replicate Python package to run Stable Diffusion models hosted on Replicate by calling replicate.run() with the model name and input prompt. Authenticate with your REPLICATE_API_TOKEN environment variable and pass the prompt to generate images easily.

Quick answer

Use the replicate Python package to run Stable Diffusion models hosted on Replicate by calling replicate.run() with the model name and input prompt. Authenticate with your REPLICATE_API_TOKEN environment variable and pass the prompt to generate images easily.

PREREQUISITES

Python 3.8+
pip install replicate
Replicate API token set as REPLICATE_API_TOKEN environment variable

Setup

Install the replicate Python package and set your Replicate API token as an environment variable for authentication.

bash

pip install replicate

output

Collecting replicate
  Downloading replicate-0.11.0-py3-none-any.whl (20 kB)
Installing collected packages: replicate
Successfully installed replicate-0.11.0

Step by step

Use the replicate.run() function to generate an image from a prompt using the Stable Diffusion model hosted on Replicate.

python

import os
import replicate

# Ensure your Replicate API token is set in the environment
# export REPLICATE_API_TOKEN="your_token_here"

model_name = "stability-ai/stable-diffusion"
prompt = "A fantasy landscape with mountains and a river at sunset"

output_urls = replicate.run(
    model_name,
    input={"prompt": prompt, "num_outputs": 1, "width": 512, "height": 512}
)

print("Generated image URL:", output_urls[0])

output

Generated image URL: https://replicate.delivery/pbxt/abc123xyz456/image.png

Common variations

Adjust image size by changing width and height in the input dictionary.
Generate multiple images by increasing num_outputs.
Use asynchronous calls with replicate.async_run() for non-blocking execution.

python

import asyncio
import os
import replicate

async def generate_images():
    model_name = "stability-ai/stable-diffusion"
    prompt = "A futuristic cityscape at night"

    output_urls = await replicate.async_run(
        model_name,
        input={"prompt": prompt, "num_outputs": 2, "width": 512, "height": 512}
    )

    for url in output_urls:
        print("Image URL:", url)

asyncio.run(generate_images())

output

Image URL: https://replicate.delivery/pbxt/def456uvw789/image1.png
Image URL: https://replicate.delivery/pbxt/def456uvw789/image2.png

Troubleshooting

If you get an authentication error, verify your REPLICATE_API_TOKEN is correctly set in your environment.
For timeout errors, try reducing image resolution or number of outputs.
If the model name is invalid, check the latest Stable Diffusion model name on Replicate's model page.

✅

Key Takeaways

Use the official replicate Python package with your API token for Stable Diffusion image generation.
Customize output images by adjusting prompt, resolution, and number of outputs in the input parameters.
Async calls with replicate.async_run() enable non-blocking image generation workflows.
Always verify your API token and model name to avoid authentication and model errors.

Verified 2026-04 · stability-ai/stable-diffusion

Verify ↗