Setup & Installation
npx skills add https://github.com/veniceai/skills --skill venice-embeddings
https://github.com/veniceai/skills/tree/main/skills/venice-embeddings
View on GitHubWhat This Skill Does
Generates vector embeddings through Venice's API using an OpenAI-compatible endpoint. Supports batch embedding, multiple models, base64 encoding, and response compression. Works with the OpenAI SDK by just changing the base URL.
The OpenAI SDK compatibility means you can swap in Venice embeddings without rewriting your pipeline, and the skill covers batch patterns, compression, and model selection so you don't have to piece it together from API docs.
When to use it
- Building a RAG pipeline with Venice-hosted embedding models
- Batch-embedding documents for semantic search in a vector database
- Switching from OpenAI embeddings to Venice with minimal code changes
- Clustering support tickets by topic using text embeddings
- Deduplicating content by comparing cosine similarity of embedded text