Open
Description
Scope check
- This is core LLM communication (not application logic)
- This benefits most users (not just my use case)
- This can't be solved in application code with current RubyLLM
- I read the Contributing Guide
Due diligence
- I searched existing issues
- I checked the documentation
What problem does this solve?
Add missing support for passing video files to LLMs that support it
Proposed solution
- Add the mime type
- Update docs
- Update
capabilities
andmodalities
code
Regarding capabilities
and modalities
: Not sure whether the existing supports_vision?
already fully covers video support for all models. Happy to look into it, if that makes sense.
Why this belongs in RubyLLM
Video is a core modality for modern AI models and user applications. Supporting video attachments in RubyLLM ensures the library remains comprehensive, relevant, and aligned with the capabilities of leading AI providers, making it easier for users to build rich, multi-modal AI solutions in Ruby.