ComfyUI Documentation

1. Project Overview & Quickstart (Comfy-Org/ComfyUI) Copy Chapter

File: README.md Copy Chapter

ComfyUI is the AI creation engine for visual professionals who demand control over every model, every parameter, and every output. Its powerful and modular node graph interface empowers creatives to generate images, videos, 3D models, audio, and more...

• ComfyUI natively supports the latest open-source state of the art models.
• Partner nodes provide access to the best closed source models such as Nano Banana, Seedance, Hunyuan3D, etc.
• It is available on Windows, Linux, and macOS, locally with our desktop application, our portable install or on our cloud.
• The most sophisticated workflows can be exposed through a simple UI thanks to App Mode.
• It integrates seamlessly into production pipelines with our API endpoints.

Get Started Copy Chapter

Local Copy Section

[Desktop Application](https://www.comfy.org/download)

• The easiest way to get started.
• Available on Windows & macOS.

[Windows Portable Package](#installing)

• Get the latest commits and completely portable.
• Available on Windows.

[Manual Install](#manual-install-windows-linux)

Supports all operating systems and GPU types (NVIDIA, AMD, Intel, Apple Silicon, Ascend).

Cloud Copy Section

[Comfy Cloud](https://www.comfy.org/cloud)

• Our official paid cloud version for those who can't afford local hardware.

Examples Copy Chapter

See what ComfyUI can do with the newer template workflows or old example workflows.

Features Copy Chapter

• A visual node graph for building and reusing image, video, audio, 3D, and text workflows without code.
• Reusable subgraphs, workflow templates, App Mode, and a local API for integrating workflows into applications.
• Efficient local execution with asynchronous queueing, partial graph re-execution, smart VRAM and RAM management, model offloading, and support for quantized models.
• Broad native model support. This is a representative list; browse the workflow library for maintained, ready-to-run templates.
  • Image generation: Stable Diffusion 1.5, SDXL, SD3.5, Flux.1, Flux.2, Qwen Image, Z-Image, Hunyuan Image 2.1, HiDream, Lumina Image 2.0, Chroma, Anima, LongCat Image, Ideogram 4, Krea 2, MageFlow, Microsoft Lens, PixelDiT, Kandinsky 5, and Ernie Image.
  • Image editing: Flux Kontext, Flux.2 Klein, Qwen Image Edit, HiDream E1.1 and O1, OmniGen2, Boogu, JoyImage Edit, MageFlow Edit, and LongCat Image Edit.
  • Video generation: Wan 2.1 and 2.2, LTX-Video 2 and 2.3, HunyuanVideo 1.5, Kandinsky 5 Video, CogVideoX, Cosmos Predict2, Bernini-R, SCAIL 2, and Mochi.
  • Audio and video generation: MiniMax H3 and LTX-AV.
  • Audio generation: ACE-Step 1.5, Stable Audio 3 and MiniMax Music 3
  • 3D and vision: Hunyuan3D 2.1, TripoSplat, SeedVR2, SUPIR, Depth Anything 3, MoGe, SAM 3 and 3.1, RT-DETRv4, and BiRefNet.
  • Text generation: Gemma 3 and 4, Qwen3, Qwen3.5, and Qwen3-VL, including multimodal inputs.
• Load complete checkpoints or separate diffusion models, VAEs, text encoders, LoRAs, ControlNets, adapters, and upscalers from supported model formats.
• Built-in tools for inpainting, outpainting, reference conditioning, masks and compositing, model merging, upscaling, frame interpolation, segmentation, depth estimation, and media processing.
• Save and load workflows as JSON, or recover complete workflows and seeds from supported generated media.
• Runs fully offline: core does not download anything unless you request it. Use --disable-api-nodes to disable the optional paid Comfy API nodes and force all built-in functionality to stay offline.
• Extend ComfyUI with custom nodes
• Configure additional model locations with extra_model_paths.yaml.

Release Process Copy Chapter

ComfyUI follows a weekly release cycle targeting Monday but this regularly changes because of model releases or large changes to the codebase. There are three interconnected repositories:

1. ComfyUI Core

   • Releases a new major stable version (e.g., v0.7.0) roughly every 2 weeks.
   • Starting from v0.4.0 patch versions will be used for fixes backported onto the current stable release.
   • Minor versions will be used for releases off the master branch.
   • Patch versions may still be used for releases on the master branch in cases where a backport would not make sense.
   • Commits outside of the stable release tags may be very unstable and break many custom nodes.
   • Serves as the foundation for the desktop release

2. Comfy Desktop

   • Builds a new release using the latest stable core version

3. ComfyUI Frontend

   • Every 2+ weeks frontend updates are merged into the core repository
   • Features are frozen for the upcoming core release
   • Development continues for the next release cycle

Shortcuts Copy Chapter

Ctrl can also be replaced with Cmd instead for macOS users

Installing

Windows and Mac Copy Chapter

We highly recommend using the desktop app:

[Link to Download](https://comfy.org/download) Copy Section

The desktop app is the easiest and best way to use ComfyUI for new users.

Windows Portable Copy Chapter

There is a portable standalone build for Windows that should work for running on Nvidia GPUs or for running on your CPU only. It is not recommended for regular users. Regular users should use the desktop app above.

Direct link to download (nvidia)

Simply download, extract with 7-Zip or with the windows explorer on recent windows versions and run. For smaller models you normally only need to put the checkpoints (the huge ckpt/safetensors files) in: ComfyUI\models\checkpoints but many of the larger models have multiple files. Make sure to follow the instructions to know which subfolder to put them in ComfyUI\models\

If you have trouble extracting it, right click the file -> properties -> unblock

The portable above currently comes with python 3.13 and pytorch cuda 13.0. Update your Nvidia drivers if it doesn't start.

All Official Portable Downloads:

Portable for AMD GPUs

Portable for Intel GPUs

Portable for Nvidia GPUs (supports 20 series and above).

Portable for Nvidia GPUs with pytorch cuda 12.6 and python 3.12 (Supports Nvidia 10 series and older GPUs, DO NOT USE THIS ON NEWER 20 SERIES AND ABOVE GPUS).

How do I share models between another UI and ComfyUI?

See the Config file to set the search paths for models. In the standalone windows build you can find this file in the ComfyUI directory. Rename this file to extra_model_paths.yaml and edit it with your favorite text editor.

[comfy-cli](https://docs.comfy.org/comfy-cli/getting-started) Copy Chapter

You can install and start ComfyUI using comfy-cli:

pip install comfy-cli
comfy install

Manual Install (Windows, Linux) Copy Chapter

Python 3.14 works but some custom nodes may have issues. The free threaded variant works but some dependencies will enable the GIL so it's not fully supported.

Python 3.13 is very well supported. If you have trouble with some custom node dependencies on 3.13 you can try 3.12

torch 2.7 is minimally supported but using a newer version is extremely recommended. Using a cu130 or above version of pytorch is required on Nvidia 20 series and above. Some features and optimizations might only work on newer versions. We generally recommend using the latest major version of pytorch with the latest cuda version unless it is less than 2 weeks old. If your pytorch is more than 6 months old, please update it.

Instructions: Copy Section

Git clone this repo.

Put your SD checkpoints (the huge ckpt/safetensors files) in: models/checkpoints

Put your VAE in: models/vae

AMD GPUs (Linux) Copy Section

AMD users can install rocm and pytorch with pip if you don't have it already installed, this is the command to install the stable version:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm7.2

This is the command to install the nightly with ROCm 7.2 which might have some performance improvements:

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm7.2

AMD GPUs (Experimental: Windows and Linux), RDNA 3, 3.5 and 4 only. Copy Section

These have less hardware support than the builds above but they work on windows. You also need to install the pytorch version specific to your hardware.

RDNA 3 (RX 7000 series):

pip install --pre torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2/gfx110X-all/

RDNA 3.5 (Strix halo/Ryzen AI Max+ 365):

pip install --pre torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2/gfx1151/

RDNA 4 (RX 9000 series):

pip install --pre torch torchvision torchaudio --index-url https://rocm.nightlies.amd.com/v2/gfx120X-all/

Intel GPUs (Windows and Linux) Copy Section

Intel Arc GPU users can install native PyTorch with torch.xpu support using pip. More information can be found here

1. To install PyTorch xpu, use the following command:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/xpu

This is the command to install the Pytorch xpu nightly which might have some performance improvements:

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu

NVIDIA Copy Section

Nvidia users should install stable pytorch using this command:

pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu130

This is the command to install pytorch nightly instead which might have performance improvements.

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu132

Troubleshooting

If you get the "Torch not compiled with CUDA enabled" error, uninstall torch with:

pip uninstall torch

And install it again with the command above.

Dependencies Copy Section

Install the dependencies by opening your terminal inside the ComfyUI folder and:

pip install -r requirements.txt

After this you should have everything installed and can proceed to running ComfyUI.

Others: Copy Section

Apple Mac silicon

You can install ComfyUI in Apple Mac silicon (M1, M2, M3 or M4) with any recent macOS version.

1. Install pytorch nightly. For instructions, read the Accelerated PyTorch training on Mac Apple Developer guide (make sure to install the latest pytorch nightly).
2. Follow the ComfyUI manual installation instructions for Windows and Linux.
3. Install the ComfyUI dependencies. If you have another Stable Diffusion UI you might be able to reuse the dependencies.
4. Launch ComfyUI by running python main.py

Note: Remember to add your models, VAE, LoRAs etc. to the corresponding Comfy folders, as discussed in ComfyUI manual installation.

Ascend NPUs

For models compatible with Ascend Extension for PyTorch (torch_npu). To get started, ensure your environment meets the prerequisites outlined on the installation page. Here's a step-by-step guide tailored to your platform and installation method:

1. Begin by installing the recommended or newer kernel version for Linux as specified in the Installation page of torch-npu, if necessary.
2. Proceed with the installation of Ascend Basekit, which includes the driver, firmware, and CANN, following the instructions provided for your specific platform.
3. Next, install the necessary packages for torch-npu by adhering to the platform-specific instructions on the Installation page.
4. Finally, adhere to the ComfyUI manual installation guide for Linux. Once all components are installed, you can run ComfyUI as described earlier.

Cambricon MLUs

For models compatible with Cambricon Extension for PyTorch (torch_mlu). Here's a step-by-step guide tailored to your platform and installation method:

1. Install the Cambricon CNToolkit by adhering to the platform-specific instructions on the Installation
2. Next, install the PyTorch(torch_mlu) following the instructions on the Installation
3. Launch ComfyUI by running python main.py

Iluvatar Corex

For models compatible with Iluvatar Extension for PyTorch. Here's a step-by-step guide tailored to your platform and installation method:

1. Install the Iluvatar Corex Toolkit by adhering to the platform-specific instructions on the Installation
2. Launch ComfyUI by running python main.py

[ComfyUI-Manager](https://github.com/Comfy-Org/ComfyUI-Manager/tree/manager-v4) Copy Chapter

ComfyUI-Manager is an extension that allows you to easily install, update, and manage custom nodes for ComfyUI.

Setup Copy Section

1. Install the manager dependencies:

   bash Copy

   pip install -r manager_requirements.txt

2. Enable the manager with the --enable-manager flag when running ComfyUI:

   bash Copy

   python main.py --enable-manager

Command Line Options Copy Section

Running

python main.py

For AMD cards not officially supported by ROCm Copy Section

Try running it with this command if you have issues:

For 6700, 6600 and maybe other RDNA2 or older: HSA_OVERRIDE_GFX_VERSION=10.3.0 python main.py

For AMD 7600 and maybe other RDNA3 cards: HSA_OVERRIDE_GFX_VERSION=11.0.0 python main.py

AMD ROCm Tips Copy Section

You can try setting this env variable PYTORCH_TUNABLEOP_ENABLED=1 which might speed things up at the cost of a very slow initial run.

Notes

Only parts of the graph that have an output with all the correct inputs will be executed.

Only parts of the graph that change from each execution to the next will be executed, if you submit the same graph twice only the first will be executed. If you change the last part of the graph only the part you changed and the part that depends on it will be executed.

Dragging a generated png on the webpage or loading one will give you the full workflow including seeds that were used to create it.

You can use () to change emphasis of a word or phrase like: (good code:1.2) or (bad code:0.8). The default emphasis for () is 1.1. To use () characters in your actual prompt escape them like \( or \).

You can use {day|night}, for wildcard/dynamic prompts. With this syntax "{wild|card|test}" will be randomly replaced by either "wild", "card" or "test" by the frontend every time you queue the prompt. To use {} characters in your actual prompt escape them like: \{ or \}.

Dynamic prompts also support C-style comments, like // comment or /* comment */.

To use a textual inversion concepts/embeddings in a text prompt put them in the models/embeddings directory and use them in the CLIPTextEncode node like this (you can omit the .pt extension):

embedding:embedding_filename.pt

How to show high-quality previews? Copy Chapter

Use --preview-method auto to enable previews.

The default installation includes a fast latent preview method that's low-resolution. To enable higher-quality previews with TAESD, download the taesd_decoder.pth, taesdxl_decoder.pth, taesd3_decoder.pth and taef1_decoder.pth and place them in the models/vae_approx folder. Once they're installed, restart ComfyUI and launch it with --preview-method taesd to enable high-quality previews.

How to use TLS/SSL? Copy Chapter

Generate a self-signed certificate (not appropriate for shared/production use) and key by running the command: openssl req -x509 -newkey rsa:4096 -keyout key.pem -out cert.pem -sha256 -days 3650 -nodes -subj "/C=XX/ST=StateName/L=CityName/O=CompanyName/OU=CompanySectionName/CN=CommonNameOrHostname"

Use --tls-keyfile key.pem --tls-certfile cert.pem to enable TLS/SSL, the app will now be accessible with https://... instead of http://....

Note: Windows users can use alexisrolland/docker-openssl or one of the 3rd party binary distributions to run the command example above.
If you use a container, note that the volume mount -v can be a relative path so ... -v ".\:/openssl-certs" ... would create the key & cert files in the current directory of your command prompt or powershell terminal.

Support and dev channel Copy Chapter

Discord: Try the #help or #feedback channels.

Matrix space: #comfyui_space:matrix.org (it's like discord but open source).

See also: https://www.comfy.org/

psst — we're hiring! Help build ComfyUI: comfy.org/careers

Frontend Development Copy Chapter

As of August 15, 2024, we have transitioned to a new frontend, which is now hosted in a separate repository: ComfyUI Frontend. The compiled JS files (from TS/Vue) are published to pypi and installed as a dependency in ComfyUI.

Reporting Issues and Requesting Features Copy Section

For any bugs, issues, or feature requests related to the frontend, please use the ComfyUI Frontend repository. This will help us manage and address frontend-specific concerns more efficiently.

Using the Latest Frontend Copy Section

The new frontend is now the default for ComfyUI. However, please note:

1. The frontend in the main ComfyUI repository is updated fortnightly.
2. Daily releases are available in the separate frontend repository.

To use the most up-to-date frontend version:

1. For the latest daily release, launch ComfyUI with this command line argument:

   text Copy

   --front-end-version Comfy-Org/ComfyUI_frontend@latest

2. For a specific version, replace latest with the desired version number:

   text Copy

   --front-end-version Comfy-Org/[email protected]

This approach allows you to easily switch between the stable fortnightly release and the cutting-edge daily updates, or even specific versions for testing purposes.

QA

Which GPU should I buy for this? Copy Section

See this page for some recommendations

File: alembic_db/README.md Copy Chapter

Generate new revision Copy Chapter

1. Update models in /app/database/models.py
2. Run alembic revision --autogenerate -m "{your message}"

File: api_server/routes/internal/README.md Copy Chapter

ComfyUI Internal Routes

All routes under the /internal path are designated for internal use by ComfyUI only. These routes are not intended for use by external applications may change at any time without notice.

File: comfy/comfy_types/README.md Copy Chapter

Comfy Typing

Type hinting for ComfyUI Node development Copy Chapter

This module provides type hinting and concrete convenience types for node developers.
If cloned to the custom_nodes directory of ComfyUI, types can be imported using:

from comfy.comfy_types import IO, ComfyNodeABC, CheckLazyMixin

class ExampleNode(ComfyNodeABC):
    @classmethod
    def INPUT_TYPES(s) -> InputTypeDict:
        return {"required": {}}

Full example is in examples/example_nodes.py.

Types

A few primary types are documented below. More complete information is available via the docstrings on each type.

`IO` Copy Chapter

A string enum of built-in and a few custom data types. Includes the following special types and their requisite plumbing:

• ANY: "*"
• NUMBER: "FLOAT,INT"
• PRIMITIVE: "STRING,FLOAT,INT,BOOLEAN"

`ComfyNodeABC` Copy Chapter

An abstract base class for nodes, offering type-hinting / autocomplete, and somewhat-alright docstrings.

Type hinting for `INPUT_TYPES` Copy Section

`INPUT_TYPES` return dict Copy Section

Options for individual inputs Copy Section

File: tests-unit/README.md Copy Chapter

Pytest Unit Tests

Install test dependencies Copy Chapter

pip install -r tests-unit/requirements.txt

Run tests Copy Chapter

pytest tests-unit/

2. Official Technical Reference & Guides (Comfy-Org/docs) Copy Chapter

ComfyUI Documentation

| English | 中文 | 日本語 | 한국어 |

Development Copy Chapter

To preview documentation changes locally, first install dependencies and then start the development server:

npm i
npm run dev

To sync translations after editing English docs, see Automated translation below (npm run translate).

Create a PR Copy Section

Create a PR. Once it is accepted Vercel will deploy the change to https://docs.comfy.org/

Generating API Reference Docs Copy Section

Can either use an OpenAPI file or URL containing the file:

cd registry/api-reference # Keep API files separated by products.
npx @mintlify/scraping@latest openapi-file

This will only generate the MDX files for each endpoint. You need to add a link to these files in docs.json, and the up-to-date API spec will be shown on that doc page.

Special Note on Renaming Files Copy Chapter

• Renaming files can cause some external links to become inaccessible, as they are already used in numerous articles and templates.
• Since we can manage the sidebar navigation can be reorganized via the docs.json file, we generally do not change the original document's file location unless absolutely necessary.
• If you renamed any files and caused the file path to be changed, please update the redirects list in the docs.json

A GitHub Action will check for redirects and fail the PR if they are missing. Redirects should follow this format:

json Copy

"redirects": [
  {
    "source": "/path/to/old-file",
    "destination": "/path/to/new-file"
  }
]

Don't forget to include the corresponding translation files under zh/, ja/, ko/, etc.!

You can also refer to the Mintlify doc to learn how to add and match a wildcard path.

About built-in node document Copy Chapter

ComfyUI now has a built-in node help menu for both built-in nodes and custom nodes. All built-in node documentation will now be maintained in this repo.

Synchronization frequency Copy Section

We will regularly sync updated documentation from the corresponding repository to docs.comfy.org weekly to ensure content synchronization and updates. If you would like to contribute to the documentation, please submit PRs and updates to this repo.

Node Documentation File Organization Copy Section

For node documentation, we will use a single-level directory structure under the built-in-node folder for the following reasons:

• ComfyUI may adjust node categories and directories during updates, using multi-level directory hierarchies means frequent adjustments to node documentation
• These frequent adjustments mean we need to frequently add redirects and checks
• Mintlify supports setting document hierarchy in the docs.json file, so we can make unified changes in that file

Due to historical updates, some existing documents use different folder hierarchies. We will no longer adjust these files, but new files will use a single-level directory

Contributing Copy Chapter

Please just create a PR and we will review it within a few days.

Or talk to us on our discord

The documentation is built with Mintlify, please refer to Mintlify documentation to learn how to use it.

i18n Contributions Copy Section

English MDX at the repo root is the source of truth. Translations mirror the same relative paths under language directories (for example zh/get_started/introduction.mdx, ja/get_started/introduction.mdx, ko/get_started/introduction.mdx). Reusable fragments live in snippets/ with per-language copies under snippets/zh/, snippets/ja/, snippets/ko/, and so on.

Contributing guides in other languages: readme/ (中文, 日本語, 한국어).

Translation policy

Supported locales are maintained through automated translation from English. When English docs change, translations are updated in batch via npm run translate — contributors do not need to hand-translate every page.

Request a new language

Want docs in another language? Open an issue with the locale you need (for example French, German, or Brazilian Portuguese). A maintainer will add the language to translation-config.json and docs.json, then run a full batch translation of all content. You only need to submit the request; no translated MDX PR is required to get started.

Specifications for editing MDX can be found in the Writing Content section of the Mintlify document.

Note: built-in-nodes/ is maintained in embedded-docs and is skipped by the translation script. Do not run bulk translation against that folder.

Automated translation

This repo includes a hash-based translation script. It compares each English file against translationSourceHash stored in the translated file; when the English source changes, the file is fully re-translated.

Prerequisites

1. Install Bun
2. Copy the env template and add your API key:

cp .env.local.example .env.local
# Set TRANSLATE_API_KEY (OpenAI-compatible: DashScope Qwen-MT, OpenRouter, DeepSeek, etc.)

npm scripts

Pass extra flags after --:

npm run translate -- --lang zh,ja
npm run translate:dry-run -- --lang ja
npm run translate -- installation/manual_install.mdx
npm run translate:check-truncation -- --lang ko
npm run translate:repair-truncated -- --lang ko

Truncated translations

Long files can occasionally be cut off mid-translation (e.g. unclosed code fences). After a batch run, the script scans newly translated files and writes a repair list to .github/i18n-logs/translate/truncation-issues.json and truncation-issues.txt (gitignored). To scan everything for a language, or to repair:

npm run translate:check-truncation -- --lang ko
npm run translate:repair-truncated -- --lang ko

repair-truncated reads the JSON log and force re-translates only the flagged files.

How it works

• Input: English MDX (primary) + existing target-language file as context (if present)
• Output: Updated files under zh/, ja/, ko/, etc., with refreshed translationSourceHash in frontmatter (snippets use an HTML comment for the hash)
• Review notes (mismatch): When the model reports semantic issues via === MISMATCHES ===, they go to .github/i18n-logs/translate/mismatches.json and mismatches.txt (gitignored), not into MDX. Only produced during npm run translate, not by the truncation scanner.
• Truncation log: Structural issues (unclosed code fences, short body) go to .github/i18n-logs/translate/truncation-issues.json — see Truncated translations above.
• Skipped paths: built-in-nodes/ (configured in translation-config.json → skip_paths)
• Chunked files: changelog/index.mdx is handled by `` version labels. The script compares EN vs target labels, translates only missing versions, and inserts them in EN order. Old blocks are never re-translated unless you use --force.
• Directories: Subdirectories are created automatically when files are written; you do not need to mkdir by hand

Script location: .github/scripts/i18n/ (see translate-i18n.ts, translation-config.json, and the i18n README for full details)

Terminology consistency

To keep the same English term rendered the same way across pages (e.g. avoid "custom node" drifting between two different Korean translations), three complementary mechanisms feed the translator. Each handles a different kind of term:

The ComfyUI frontend (ComfyUI_frontend/src/locales) is the authoritative source of term translations. npm run glossary:sync mirrors its locale terms into glossary/frontend/{lang}.json (rebuilt wholesale each run — never hand-edit). Hand-maintained corrections live in glossary/overrides/{lang}.json, which wins over the mirror and is the place to record a term decision or drop a noisy frontend term:

// glossary/overrides/ko.json
{
  "terms":  { "custom node": "커스텀 노드" },   // remap or add (wins over frontend)
  "ignore": ["title", "additional", "work"]      // drop a noisy frontend term
}

At translation time, only terms that actually appear in a document are selected and injected as preferred (not mandatory) hints, so the model keeps natural phrasing when a literal substitution would read awkwardly. ComfyUI proper nouns with no settled translation (model names, checkpoint, …) instead go in preserve_terms to stay in English. See the i18n README for the full design and curation guidance.

npm run glossary:sync                 # rebuild the frontend mirror, all languages
npm run glossary:sync -- --lang ko    # one language
npm run glossary:sync:dry-run         # report counts without writing

The frontend locales source resolves in order:

• Remote (default): frontend_locales_url in translation-config.json (GitHub raw main branch). Override with FRONTEND_LOCALES_URL or --frontend-url .
• Local (optional): --frontend or FRONTEND_LOCALES_PATH when you need an offline or forked checkout.

Quality review

npm run translate:review scores existing translations with an independent (and typically cheaper) AI model — an LLM-as-a-judge — on four axes: accuracy, completeness, terminology (checked against the glossary), and fluency. It is separate from the translation model's own === MISMATCHES === self-notes; here a different model acts as judge.

Results are advisory: the detailed scores and issue list are written to .github/i18n-logs/review/ (quality-report.json / .txt, gitignored) and never block a PR. The reviewed hash is recorded as reviewSourceHash in the translated file's frontmatter (committed to git), so review state is shared across the team and visible per file — mirroring translationSourceHash. By default only translations that are up to date with English and not yet reviewed at that hash are checked.

npm run translate:review                     # pending reviews, all languages
npm run translate:review -- --lang ko        # one language
npm run translate:review -- --all            # re-review everything
npm run translate:review -- --sample 20      # N pending files per language
npm run translate:review -- --min-score 4    # report files scoring below 4/5

Configure a dedicated cheap judge model via REVIEW_API_KEY / REVIEW_API_BASE_URL / REVIEW_API_MODEL in .env.local (falls back to the TRANSLATE_* model when unset). Use a fast model — evaluation is lighter than translation; reasoning-heavy models are slow under concurrency, so lower REVIEW_CONCURRENCY if you see connection errors.

Adding a new language

See Request a new language above — please open an issue rather than adding a language in a PR yourself.

Maintainers: add one entry under languages in .github/scripts/i18n/translation-config.json (code, name, dir, snippets_dir). Path exclusion, link localization, and English-file scanning are derived automatically by i18n-config.mjs in the same folder — no per-language edits in translate scripts when adding a locale. Then add navigation in docs.json (see Mintlify Localization), and batch-translate:

npm run glossary:sync -- --lang fr        # build the terminology glossary for the new locale
npm run translate:dry-run -- --lang fr
npm run translate -- --lang fr
npm run translate:snippets -- --lang fr

The glossary is extended automatically: once the language is in translation-config.json, npm run glossary:sync generates glossary/frontend/{lang}.json from the ComfyUI frontend locale (provided the frontend ships that locale). An glossary/overrides/{lang}.json file is optional — add it later only to correct or pin specific terms. See Terminology consistency.

Manual translation

You can still translate by hand instead of using the script:

1. Create a file under the language directory with the same filename and path as the English original.
2. Localize import paths (/snippets/... → /snippets/zh/...) and internal links (/path → /zh/path).
3. Add the page path to the correct language group in docs.json.

When English MDX changes, the i18n-sync-check workflow warns if matching translation files were not updated in the same PR, and posts a reminder comment tagging @comfyui-wiki. Update translations manually or re-run npm run translate.

Contributing Workflow Examples

When adding workflow examples to the documentation, please:

1. Take your output from ComfyUI (PNG, WebP), and add the models urls to the workflow so a user will have them when they drag in the workflow. You can use this tool to edit the metadata of the PNG or WebP file.

2. Upload your workflow JSON and preview image to the example_workflows repository

3. Use the raw GitHub content URL in your documentation. To convert a GitHub file URL to a raw content URL:

   • Start with your GitHub file URL:
     text Copy

     https://github.com/Comfy-Org/example_workflows/blob/main/your-workflow.json

   • Change it to raw.githubusercontent.com and remove '/blob':
     text Copy

     https://raw.githubusercontent.com/Comfy-Org/example_workflows/main/your-workflow.json

   You can also click the "Raw" button on the GitHub file page and copy the URL directly.

This ensures the workflow metadata is preserved in the docs site when dragging it into ComfyUI.