Extensions/ComfyUI-Replicate
ComfyUI Extension

ComfyUI-Replicate

Run [a/Replicate models](https://replicate.com/explore) in ComfyUI.

By replicate·Created 2 years ago·Updated 2 years ago· 207
replicate/comfyui-replicate
Nodes62
On cloudLocal install
CategoryReplicate
Stars207
Updated2 years ago

Nodes (62)

Replicate ai-forever/kandinsky-2

The two-stage text-to-image model, via API

Replicate
Replicate ai-forever/kandinsky-2.2

Run Kandinsky 2.2 in ComfyUI

Replicate
Replicate andreasjansson/blip-2

Caption an image or ask it a question

Replicate
Replicate andreasjansson/deepfloyd-if

The pixel-space text-rendering model, via API

Replicate
Replicate batouresearch/high-resolution-controlnet-tile

Creative tile upscaling in ComfyUI

Replicate
Replicate batouresearch/magic-image-refiner

Refine and enhance an image

Replicate
Replicate batouresearch/magic-style-transfer

IP-Adapter style transfer on SDXL

Replicate
Replicate black-forest-labs/flux-1.1-pro

The faster, better top Flux tier

Replicate
Replicate black-forest-labs/flux-dev

The workhorse Flux tier via API

Replicate
Replicate black-forest-labs/flux-pro

The API-only top Flux tier in ComfyUI

Replicate
Replicate black-forest-labs/flux-schnell

The fast, cheap Flux tier via API

Replicate
Replicate bytedance/sdxl-lightning-4step

4-step SDXL, run in the cloud

Replicate
Replicate cjwbw/hyper-sdxl-1step-t2i

One-step SDXL image generation

Replicate
Replicate cjwbw/pixart-sigma

PixArt-Σ text-to-image from a ComfyUI node

Replicate
Replicate cjwbw/seamless_communication

Meta's SeamlessM4T translation node

Replicate
Replicate cjwbw/supir

Restoration-grade upscaling without the VRAM bill

Replicate
Replicate cuuupid/glm-4v-9b

A modern vision-language model for image Q&A

Replicate
Replicate cuuupid/idm-vton

Virtual try-on (put a garment on a person) in ComfyUI

Replicate
Replicate declare-lab/tango

Text-to-audio sound effects in ComfyUI

Replicate
Replicate fofr/consistent-character

One reference into many poses

Replicate
Replicate fofr/controlnet-preprocessors

Run every ControlNet preprocessor at once

Replicate
Replicate fofr/face-to-many

Turn a face into 3D, emoji, clay, or pixel art

Replicate
Replicate fofr/latent-consistency-model

Fast LCM generation with ControlNet

Replicate
Replicate fofr/sd3-with-chaos

SD3 with a randomness dial, run via API

Replicate
Replicate fofr/style-transfer

Copy a style image's look onto a new generation

Replicate
Replicate fofr/ultrapixel

Ultra-high-res image synthesis via API

Replicate
Replicate ideogram-ai/ideogram-v2

The text-in-image model, via API

Replicate
Replicate ideogram-ai/ideogram-v2-turbo

Best-in-class text in images, fast

Replicate
Replicate lucataco/hunyuandit-v1.1

Tencent's bilingual DiT via API

Replicate
Replicate lucataco/llama-3-vision-alpha

Llama 3 with eyes, via API

Replicate
Replicate lucataco/magnet

Meta's MAGNeT text-to-music in ComfyUI

Replicate
Replicate lucataco/moondream2

The tiny fast image captioner, via API

Replicate
Replicate lucataco/paligemma-3b-pt-224

Google's VLM, and the prompt syntax you need to know

Replicate
Replicate lucataco/pasd-magnify

PASD detail-adding upscaler in ComfyUI

Replicate
Replicate lucataco/qwen-vl-chat

Alibaba's VLM, strong on text-in-image

Replicate
Replicate lucataco/sdxl-clip-interrogator

Reverse-engineer a prompt from an image

Replicate
Replicate lucataco/xtts-v2

Clone a voice and read text in it, from ComfyUI

Replicate
Replicate luosiallen/latent-consistency-model

The original LCM, run via API

Replicate
Replicate meta/llama-2-70b-chat

The older, bigger Llama chat node

Replicate
Replicate meta/llama-2-7b-chat

The cheap, fast Llama 2 node

Replicate
Replicate meta/meta-llama-3.1-405b-instruct

Meta's biggest LLM as a ComfyUI node

Replicate
Replicate meta/meta-llama-3-70b-instruct

The practical LLM node for prompt work

Replicate
Replicate meta/meta-llama-3-8b-instruct

Run an LLM inside your ComfyUI graph

Replicate
Replicate meta/musicgen

Text-to-music generation in ComfyUI

Replicate
Replicate nateraw/audio-super-resolution

Upscale audio quality via API

Replicate
Replicate nateraw/musicgen-songstarter-v0.2

AI-generated loops for producers

Replicate
Replicate omniedgeio/face-swap

One-shot face swapping in ComfyUI

Replicate
Replicate pharmapsychotic/clip-interrogator

Reverse an image into a prompt

Replicate
Replicate philz1337x/clarity-upscaler

The open Magnific-style upscaler in ComfyUI

Replicate
Replicate recraft-ai/recraft-v3

The design-focused #1 leaderboard model via API

Replicate
Replicate salesforce/blip

Image captioning and visual Q&A in ComfyUI

Replicate
Replicate smoretalk/rembg-enhance

One-click background removal via API

Replicate
Replicate snowflake/snowflake-arctic-instruct

The enterprise LLM node with an unusual schema

Replicate
Replicate stability-ai/sdxl

Run SDXL (text-to-image, img2img, inpaint) via API

Replicate
Replicate stability-ai/stable-diffusion

The original SD, run via API

Replicate
Replicate stability-ai/stable-diffusion-3

SD3 via API, text and img2img

Replicate
Replicate suno-ai/bark

Run Bark text-to-audio from inside ComfyUI

Replicate
Replicate tstramer/material-diffusion

Tileable textures via API

Replicate
Replicate yorickvp/llava-13b

The classic image captioner, via API

Replicate
Replicate yorickvp/llava-v1.6-34b

The strongest LLaVA node in this pack

Replicate
Replicate yorickvp/llava-v1.6-mistral-7b

The fast, cheap LLaVA-NeXT node

Replicate
Replicate zsxkib/realistic-voice-cloning

RVC song covers in ComfyUI

Replicate
Readme

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.

example-screenshot

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.