Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
This extension can translate prompt from your native language into English, so you can write prompt with your native language
| Date | Stars |
|---|---|
| 2026-07-31 | 252 |
| 2026-08-05 | 252 |
| 2026-08-06 | 252 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
### Language
[中文](README.cn.md)
# Stable-Diffusion-Webui-Prompt-Translator
This Stable-Diffusion-Webui's extension can translate prompt from your native language into English, so from now on, you can write prompt with your native language.
# Install
GO to SD webui's extension page, choose install from github url. Fill it with this project's url, click install.
(If you can not connect to github, you can just download this project as `.zip` file. Unzip it to `Your Stable-Diffusion-Webui/extensions` folder.)
**Then, Re-Launch Stable-Diffusion-Webui, Just reload UI won't work.**
# How to use
## Set translation API Key
This extension uses Online AI translator service [Deepl.com](https://www.deepl.com), [Google](https://cloud.google.com/translate/docs/setup?hl=en), [Yandex](https://yandex.com/dev/translate/) and [Baidu](http://api.fanyi.baidu.com/)(for Chinese)'s API to translate your pormpt.
Or, use local AI translation engine [MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)
For online service, you need go to those translation website, apply a free API Key for this. Fill your API Key to this extension's tab page, click save, done.
(For baidu, check [README.cn.md](README.cn.md))
For local AI translation engine MarianMT, you need to do a lot of downloading work, and the quality is not as good as online service. It also need to load a model when using. But if you still want it, check [Setup for MarianMT](#setup-for-marianmt) section.
### Setup for Deepl
With deepl, it offers you 500,000 character/month for free. That's more than enough.
After applying the API, go to your acount, the API key is at the bottom like following:
After you get your APP Key, fill it into this extension's tab page, then click save.
### Setup for Yandex
On Yandex's translation API service's page:
[https://yandex.com/dev/translate/](https://yandex.com/dev/translate/)
There is a button for "Get Free API Key" at Step 3.
Do that to get your APP ID and APP Key, then fill them into this extension.
With Yandex, APP ID is called `Folder ID`, APP Key is called `IAM token`.
I don't know the detail for using Yandex. I don't use yandex. The code for yandex is provided by pull request and the code looks good so I merged it. But no document for that. You need to try it yourself.
### Setup for Google translation API Key [check](https://cloud.google.com/translate/docs/setup?hl=en)
**Note: Google API does not offer any free words per month, and setup is kind of complex for normal user. Only choose google when you can not use other service.**
1. create GCP project([check](https://cloud.google.com/resource-manager/docs/creating-managing-projects?hl=en))
2. enable "Cloud Translation API"
- search "translation"
- click "Cloud Translation API"
- click "ENABLE"
3. create "API Key"
- after "Cloud Translation API" click "MANAGE"
- select PROEJCT, and godo "API/Service Details"
- click "CREDENTIALS" tab and click "CREATED CREDENTIALS"
If you are in China, just choose [Baidu](http://api.fanyi.baidu.com/).
### Setup for MarianMT
[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian) is an open source local AI translation engine. But the quality is not as good as online translation services.
To use it, you need do following things:
Go to extension folder, create a "MarianMT" Subfolder.
#### Download MarianMT Model
Go to [MarianMT's model page](https://huggingface.co/Helsinki-NLP). There are 1000+ models in the list.
Each model is target on a Source languge to a Target languge task.
For example, model `opus-mt-en-ja` means this model can translate English to Japanese. And `opus-mt-ja-en` means it can translate Japanese to English.
So, you need 2 models for your native language.
Go to a model's project page, you will find,Excerpt of 7,497 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:c84cf8d17b319e97, name:stable diffusion