AI Long Text Translation Tool

An AI large language model–based smart long-text translation tool that supports paragraph-integrity splitting, concurrent translation, and batch file processing.

Bug report

Loading tool, please wait...

If the online tool fails to load or run, try the desktop tool.https://tools.yikeaigc.com/

Tool Usage

Return to old version

API Domain

API Key

Model Name

Target Language

Concurrency

Text to Translate

Text will be split by complete paragraphs, max 4000 characters each

API Domain

API Key

Model Name

Target Language

Concurrency

Choose TXT file

Select multiple TXT files for batch translation; each file will be split by complete paragraphs

Instructions

Software Usage Instructions

Configure API information: Enter the OpenAI-compatible API domain, key, and model name. Custom API providers are supported; make sure the API format is consistent with OpenAI.
Select the target language:
- Preset languages: Choose from 10 major global languages, including English, Chinese, Spanish, French, German, Japanese, Korean, Russian, Arabic, and Portuguese
- Custom language: Select the "Custom Language" option and enter the specific language name in the pop-up input box
Set the concurrency count: Select a concurrency count from 1 to 20 based on API limits and network conditions. It is recommended to start testing with a smaller value.
Single-text translation mode:
- Enter or paste the content to be translated in the text box
- The system automatically splits the text by paragraph integrity, with each segment not exceeding 4000 characters
- Click the "Start Translation" button to view translation progress in real time
- After translation is complete, you can copy the results or download a TXT file
Batch file translation mode:
- Select multiple TXT files for batch upload
- The system displays information for the first 20 files, and the remaining files will be processed as well
- Each file is processed independently, with detailed processing progress displayed
- After all files are complete, click "Download All Results (ZIP)" to get the compressed package
View results: In single-text mode, you can directly view and copy the results. In batch mode, a ZIP archive download is provided, containing all translated files.

FAQ

A: All providers that are compatible with the OpenAI API format are supported, including official OpenAI, Azure OpenAI, major domestic AI service providers, and more. As long as the API interface format is consistent and the correct domain and key are configured, it can be used.

A: The system prioritizes splitting text by paragraph integrity, with each segment not exceeding 4000 characters. If a single paragraph is too long, it will be split a second time by sentence to ensure translation coherence and accuracy. The splitting process fully preserves the original formatting structure.

A: The concurrency level depends on the API provider’s rate limits and network conditions. We recommend starting with 3. If the API supports higher concurrency and the network is stable, you can gradually increase it to 5-10. Excessive concurrency may cause API rate limiting or request failures.

A: No. All file processing is done locally on your device, and the text content is only sent to the AI service provider you have configured for translation. The tool does not upload any files to our servers, ensuring your data privacy and security.

A: Translation quality mainly depends on the AI model and text segmentation strategy you choose. We recommend using GPT-3.5-turbo or a more advanced model. The system’s intelligent segmentation ensures contextual coherence and delivers professional-grade translation results.

A: The system will display the processing status of each file, and failed files will be marked with the error reason. Successfully translated files can still be downloaded normally. Common causes of failure include API key errors, network timeouts, file format issues, etc. Please check your configuration and reprocess the failed files.

A: Currently, only plain text files in TXT format are supported. UTF-8 encoding is recommended to ensure Chinese and other special characters display correctly. Please convert files in other formats to TXT before translating.