- Fixed NLLB-200 tokenizer forced_bos_token_id issue
- Changed from lang_code_to_id to convert_tokens_to_ids
- Added .dockerignore to exclude models directory from Docker build
- Prevents disk space issues during build
- Models are loaded at runtime via volume mount
- Both M2M100 and NLLB-200 models tested and working
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Added optional 'model' parameter to translation request (default: m2m100)
- M2M100: 105 languages, Apache 2.0 License (commercial OK)
- NLLB-200: 200 languages, CC-BY-NC 4.0 License (non-commercial only)
- Updated /api/translate endpoint to accept model selection
- Updated /api/supported-languages to show languages per model
- Added comprehensive language name mappings for all NLLB-200 languages
- Both models can be used independently with automatic model loading
- Model information includes license and commercial use status
Example usage:
- Default (M2M100): {"text": "Hello", "source_lang": "en", "target_lang": "ko"}
- NLLB-200: {"text": "Hello", "source_lang": "en", "target_lang": "ko", "model": "nllb200"}
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Updated API title from "Malaysian Language Translation API" to "Multilingual Translation API"
- Updated API description to mention 105+ languages and M2M100 model
- Updated /api/translate endpoint docstring to reflect multilingual support
- Updated startup/shutdown log messages
- Added commercial license note (Apache 2.0) in API description
This ensures the Swagger UI (http://localhost:8001/docs) shows correct information.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
- Implemented REST API for 105+ language translation
- Used Facebook M2M100 model (Apache 2.0 License - Commercial use allowed)
- Supports any-to-any translation between 105 languages
- Major languages: English, Chinese, Spanish, Arabic, Russian, Japanese, Korean, etc.
- Southeast Asian: Malay, Indonesian, Thai, Vietnamese, Tagalog, Burmese, Khmer, Lao
- South Asian: Bengali, Hindi, Urdu, Tamil, Telugu, Marathi, Gujarati, etc.
- European: German, French, Italian, Spanish, Portuguese, Russian, etc.
- African: Swahili, Amharic, Hausa, Igbo, Yoruba, Zulu, Xhosa
- And many more languages
Tech Stack:
- FastAPI for REST API
- Transformers (Hugging Face) for ML model
- PyTorch for inference
- Docker for containerization
- M2M100 418M parameter model
Features:
- Health check endpoint
- Supported languages listing
- Dynamic language validation
- Model caching for performance
- GPU support (auto-detection)
- CORS enabled for web clients
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
