ComfyUI-Replicate
Run [a/Replicate models](https://replicate.com/explore) in ComfyUI.
Nodes (62)
The two-stage text-to-image model, via API
Run Kandinsky 2.2 in ComfyUI
Caption an image or ask it a question
The pixel-space text-rendering model, via API
Creative tile upscaling in ComfyUI
Refine and enhance an image
IP-Adapter style transfer on SDXL
The faster, better top Flux tier
The workhorse Flux tier via API
The API-only top Flux tier in ComfyUI
The fast, cheap Flux tier via API
4-step SDXL, run in the cloud
One-step SDXL image generation
PixArt-Σ text-to-image from a ComfyUI node
Meta's SeamlessM4T translation node
Restoration-grade upscaling without the VRAM bill
A modern vision-language model for image Q&A
Virtual try-on (put a garment on a person) in ComfyUI
Text-to-audio sound effects in ComfyUI
One reference into many poses
Run every ControlNet preprocessor at once
Turn a face into 3D, emoji, clay, or pixel art
Fast LCM generation with ControlNet
SD3 with a randomness dial, run via API
Copy a style image's look onto a new generation
Ultra-high-res image synthesis via API
The text-in-image model, via API
Best-in-class text in images, fast
Tencent's bilingual DiT via API
Llama 3 with eyes, via API
Meta's MAGNeT text-to-music in ComfyUI
The tiny fast image captioner, via API
Google's VLM, and the prompt syntax you need to know
PASD detail-adding upscaler in ComfyUI
Alibaba's VLM, strong on text-in-image
Reverse-engineer a prompt from an image
Clone a voice and read text in it, from ComfyUI
The original LCM, run via API
The older, bigger Llama chat node
The cheap, fast Llama 2 node
Meta's biggest LLM as a ComfyUI node
The practical LLM node for prompt work
Run an LLM inside your ComfyUI graph
Text-to-music generation in ComfyUI
Upscale audio quality via API
AI-generated loops for producers
One-shot face swapping in ComfyUI
Reverse an image into a prompt
The open Magnific-style upscaler in ComfyUI
The design-focused #1 leaderboard model via API
Image captioning and visual Q&A in ComfyUI
One-click background removal via API
The enterprise LLM node with an unusual schema
Run SDXL (text-to-image, img2img, inpaint) via API
The original SD, run via API
SD3 via API, text and img2img
Run Bark text-to-audio from inside ComfyUI
Tileable textures via API
The classic image captioner, via API
The strongest LLaVA node in this pack
The fast, cheap LLaVA-NeXT node
RVC song covers in ComfyUI
comfyui-replicate
Custom nodes for running Replicate models in ComfyUI.
Take a look at the example workflows and the supported Replicate models to get started.
Set your Replicate API token before running
Make sure you set your REPLICATE_API_TOKEN in your environment. Get your API tokens here, we recommend creating a new one:
https://replicate.com/account/api-tokens
To pass in your API token when running ComfyUI you could do:
On MacOS or Linux:
export REPLICATE_API_TOKEN="r8_************"; python main.py
On Windows:
set REPLICATE_API_TOKEN="r8_************"; python main.py
Direct installation
cd ComfyUI/custom-nodes
git clone https://github.com/replicate/comfyui-replicate
cd comfyui-replicate
pip install -r requirements.txt
Supported Replicate models
View the supported_models.json to see which models are packaged by default.
Update Replicate models
Simply run ./import_schemas.py to update all model nodes. The latest version of a model is used by default.
Add more models
Only models that return simple text or image outputs are currently supported. If a model returns audio, video, JSON objects or a combination of outputs, the node will not work as expected.
If you want to add more models, you can:
- add the model to
supported_models.json(for example,fofr/consistent-character) - run
./import_schemas.py, this will update all schemas and import your new model - restart ComfyUI
- use the model in workflow, it’ll have the title ‘Replicate [model author/model name]’
Roadmap
Things to investigate and add to this custom node package:
- support for more types of Replicate model (audio and video first)
- showing logs, prediction status and progress (via tqdm)
Contributing
If you add models that others would find useful, feel free to raise PRs.