Qwen2 5 Vl Python Packages

jina-v4-vllm-plugin

Serve Jina Embeddings v4 multi-vector (ColBERT-style) multimodal text+image embeddings from a stock vLLM OpenAI server via an out-of-tree plugin, with configurable image fidelity.

relsim

Relational Visual Similarity - A new visual similarity notion that captures the internal relational logic of a scene