Added
- Initial release of Token Vision
- Core token calculation functionality for images
- Support for multiple AI models:
- Claude (Sonnet, Haiku, Opus)
- GPT-4o (Latest, Nov 2024, Aug 2024, May 2024)
- Gemini (Pro, Flash)
- Custom model configuration support
- Batch processing capabilities
- Cost calculation with support for:
- Standard rates
- Batch processing rates
- Cached input rates
- Comprehensive documentation and examples
- Type hints and docstrings
- Development tools configuration (black, isort, mypy, ruff)
- MIT License
Changed
- Renamed package from "image-token-calculator" to "token-vision"
- Updated model configurations to match latest pricing
- Improved token calculation accuracy
Removed
- GPT-4V model (replaced with GPT-4o)