Skip to content

Conversation

@ScReameer
Copy link

Key changes

  • Enable multimodal embeddings: text stays on embed, image requests now call image_embed, and an audio path is stubbed with NotImplementedError for future support.
  • Update infinity-emb to 0.0.77.
  • Pin optimum==1.24.0 so the Docker image no longer crashes with ModuleNotFoundError.
  • Extend handler.py to read an optional modality field (text default; accepts text, image, audio) for both native and OpenAI-compatible payloads.
  • Updated RunPod tests .runpod/tests.json to cover text and image scenarios
  • Add a repository-level httpx.AsyncClient with configurable timeout, pooling, and User-Agent to stabilise remote image downloads.
  • Add a local testing with pytest tests -v

@ScReameer ScReameer marked this pull request as ready for review December 1, 2025 14:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant