Mid-size Qwen3 cross-encoder reranker with 32K context, balancing speed and accuracy for document relevance scoring in search and RAG workloads. 4B size.
LLM Gateway routes requests to the best providers that are able to handle your prompt size and parameters.