# AI use cases for Index Translate (bilibili/Index-Translate) Repo: [bilibili/Index-Translate](https://github.com/bilibili/Index-Translate) | Use cases: 4 | Page: https://dedupedai.com/use-cases/bilibili-index-translate/ Index-Translate is a family of multilingual, multimodal translation models built on Qwen3.5, with a free OpenAI-compatible API and a command-line script. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Translate a screenshot or photo without OCR: https://dedupedai.com/use-cases/bilibili-index-translate/translate-screenshot-without-ocr/ - Translate a text string from the command line: https://dedupedai.com/use-cases/bilibili-index-translate/translate-text-from-command-line/ - Use the API inside a browser translation extension: https://dedupedai.com/use-cases/bilibili-index-translate/translate-in-browser-with-local-bridge/ - Turn speech into translated subtitles: https://dedupedai.com/use-cases/bilibili-index-translate/make-subtitles-from-speech/ ## How to translate a screenshot or photo without OCR with Index Translate Repo: [bilibili/Index-Translate](https://github.com/bilibili/Index-Translate) | Difficulty: Intermediate | Time: 15 minutes | Basis: From the README Get the translated text from an image in one command, with no separate OCR step. **Who it is for:** A researcher, student, or support agent who has a screenshot or photo in a foreign language and needs the text quickly. **The problem:** Normal translation tools need the text copied out of the image first, which is slow and often fails on layouts, diagrams, or photos. **The outcome:** A translated version of the text shown in the image, produced directly from the picture. ### Steps 1. Clone the repo with `git clone https://github.com/bilibili/Index-Translate` and open a terminal in the folder. (inferred, check the README) 2. Run `python inference/llm/call_api.py --image path/to/screenshot.png --target zh` to translate with the default 35B model. 3. For a lighter model, run `python inference/llm/call_api.py --image photo.jpg --target zh -m Index-Translate-2B`. 4. Compare the output against the original image and note any text the model missed. ### Prompt to paste into your agent ```text Use the repo at https://github.com/bilibili/Index-Translate. Run `python inference/llm/call_api.py --image --target ` on the image at and return the translated text. Verify that the output is in the target language and that it covers the main text visible in the image. ``` ### Caveat The README gives no rate limits or uptime guarantee for the free public API, so large batches may not run reliably. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/bilibili-index-translate/translate-screenshot-without-ocr/ ## How to translate a text string from the command line with Index Translate Repo: [bilibili/Index-Translate](https://github.com/bilibili/Index-Translate) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Send a sentence to the free API and get a translation back in one command. **Who it is for:** A developer or writer who needs a quick translation while working in a terminal and has no local GPU. **The problem:** Setting up a local translation model needs GPUs and model downloads, which is too much for a single short sentence. **The outcome:** A translated sentence printed in the terminal, produced by the free hosted model. ### Steps 1. Clone the repo with `git clone https://github.com/bilibili/Index-Translate` and open a terminal in the folder. (inferred, check the README) 2. Run `python inference/llm/call_api.py "Hello, world!" --target zh` to translate a sentence to Chinese. 3. Replace the sentence and the `--target` value with the text and language you need. 4. Check the output for the meaning and for any terminology that must stay fixed. ### Prompt to paste into your agent ```text Use the repo at https://github.com/bilibili/Index-Translate. Run `python inference/llm/call_api.py "" --target ` and return the translated output. Verify that the output is in the requested language and keeps the meaning of the source text. ``` ### Caveat The free API is a hosted service, so text you send leaves your machine, which may not suit confidential material. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/bilibili-index-translate/translate-text-from-command-line/ ## How to use the API inside a browser translation extension with Index Translate Repo: [bilibili/Index-Translate](https://github.com/bilibili/Index-Translate) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Run a local bridge so Immersive Translate can translate web pages through the free API. **Who it is for:** A reader who browses foreign-language pages and uses a browser extension such as Immersive Translate. **The problem:** Browser extensions cannot always reach the API directly because of CORS and WAF headers, so the extension needs a local bridge. **The outcome:** Web pages translated in the browser, with requests routed through a local address on your machine. ### Steps 1. Clone the repo with `git clone https://github.com/bilibili/Index-Translate` and open a terminal in the folder. (inferred, check the README) 2. Run `python inference/llm/call_api.py --serve` to start the local bridge proxy. 3. In Immersive Translate, set Service to Custom / OpenAI and API URL to http://127.0.0.1:8080/v1. 4. Set Model to Index-Translate-35B-A3B and API Key to any string, such as index, then translate a test page. ### Prompt to paste into your agent ```text Use the repo at https://github.com/bilibili/Index-Translate. Start the local bridge with `python inference/llm/call_api.py --serve` and confirm it listens at http://127.0.0.1:8080/v1. Return the exact extension settings used. Verify that a test request to that address returns a translated response. ``` ### Caveat The bridge only runs while the script is running on your machine, so the extension stops working when you close the terminal. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/bilibili-index-translate/translate-in-browser-with-local-bridge/ ## How to turn speech into translated subtitles with Index Translate Repo: [bilibili/Index-Translate](https://github.com/bilibili/Index-Translate) | Difficulty: Advanced | Time: 1 hour | Basis: From the README Use Index-Echo S2TT to produce subtitles from speech in zh to en, ja, or es. **Who it is for:** A video editor or creator who has spoken audio in Chinese and needs subtitles in English, Japanese, or Spanish. **The problem:** Transcribing speech and then translating the transcript is two steps, and timing often gets lost between them. **The outcome:** Subtitles translated from the speech in one pipeline, for the supported language pairs. ### Steps 1. Open the Index-Echo S2TT guide at inference/echo-s2tt/README.md. 2. Choose the model you need from the Index-Echo S2TT 2B or 9B checkpoints on Hugging Face or ModelScope. 3. Follow the packaged script in the guide for the zh to en, ja, or es pair. 4. Review the subtitles against the audio, especially names and numbers. ### Prompt to paste into your agent ```text Use the repo at https://github.com/bilibili/Index-Translate. Follow inference/echo-s2tt/README.md to translate the speech file at from zh to and return the subtitle output. Verify that the target language is one of en, ja, or es and that subtitles follow the speech order. ``` ### Caveat The README lists S2TT support only for zh to en, ja, and es, so other source or target pairs are not covered. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/bilibili-index-translate/make-subtitles-from-speech/