All committers have signed the CLA.

Thank you for your submission!

This service-oriented model for invocation, which helps zvec achieve the RAG capability, is what we currently lack.
You can implement your HttpEmbedingFunction(OllmaEmbedingFunction) by inheriting DenseEmbedingFunction in the directory python/zvec/extension. This will make it easier for more users to use!

https://zvec.org/api-reference/python/extension/#zvec.extension.DenseEmbeddingFunction

Thanks for the feedback! Moved the implementation into python/zvec/extension/http_embedding_function.py as HTTPDenseEmbedding, inheriting from DenseEmbeddingFunction. It's now exported from zvec.extension and the example imports it from there instead of defining the class inline.

@greptile

Greptile Summary

This PR adds HTTPDenseEmbedding, a stdlib-only implementation for OpenAI-compatible embedding endpoints (LM Studio, Ollama, vLLM, LocalAI). The implementation follows established project patterns including @lru_cache for result caching and the DenseEmbeddingFunction protocol.

Key additions:

• HTTPDenseEmbedding class with automatic dimension detection
• Comprehensive example with CLI interface for easy testing
• Zero external dependencies (uses only urllib.request and stdlib)
• Proper error handling for network failures and malformed responses
• Consistent with existing embedding function implementations

The code is production-ready, well-documented, and provides valuable functionality for users running local inference servers.

Confidence Score: 5/5

• This PR is safe to merge with no identified issues
• The code follows established project patterns, includes comprehensive documentation and error handling, and has been tested against live servers. No breaking changes or security concerns identified.
• No files require special attention

Important Files Changed

Sequence Diagram

sequenceDiagram
    participant User
    participant HTTPDenseEmbedding
    participant Cache
    participant Server as Local Server<br/>(LM Studio/Ollama)
    
    User->>HTTPDenseEmbedding: __init__(base_url, model)
    HTTPDenseEmbedding->>HTTPDenseEmbedding: Store config
    
    User->>HTTPDenseEmbedding: dimension property
    HTTPDenseEmbedding->>HTTPDenseEmbedding: embed("dimension probe")
    HTTPDenseEmbedding->>Server: POST /v1/embeddings
    Server-->>HTTPDenseEmbedding: {data: [{embedding: [...]}]}
    HTTPDenseEmbedding->>Cache: Store result
    HTTPDenseEmbedding-->>User: vector dimension
    
    User->>HTTPDenseEmbedding: embed("user text")
    HTTPDenseEmbedding->>Cache: Check cache
    alt Cache hit
        Cache-->>HTTPDenseEmbedding: Cached vector
    else Cache miss
        HTTPDenseEmbedding->>Server: POST /v1/embeddings
        Server-->>HTTPDenseEmbedding: {data: [{embedding: [...]}]}
        HTTPDenseEmbedding->>Cache: Store result
    end
    HTTPDenseEmbedding-->>User: vector

_{Last reviewed commit: eb3960e}

Removed the example file as requested — server-dependent examples belong in zvec-web.