netease-youdao/BCEmbedding

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1883 stars · 130 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

BCEmbedding is NetEase Youdao's open-source library of bilingual and crosslingual embedding and reranker models for English and Chinese, built for RAG applications. It provides an EmbeddingModel for semantic vectors and a RerankerModel for refining and ranking search results, with integrations for LangChain and LlamaIndex.

Use cases

• generate embeddings for semantic search in english and chinese
• rerank retrieved documents in a rag pipeline
• build a bilingual question-answering system
• integrate embedding models into langchain or llama index
• evaluate embedding quality with mteb benchmarks
• improve retrieval accuracy for crosslingual queries

When to choose

• you need strong english-chinese bilingual or crosslingual retrieval for a RAG product
• you want both embedding and reranker models from one consistent, benchmarked family
• you use LangChain or LlamaIndex and want drop-in model integrations

When to avoid

• you need multilingual coverage beyond English and Chinese
• you need a lightweight model for edge or CPU-only deployment without GPU resources
• you only need a general-purpose embedding model with no reranking

Facets

library · maturity active

machine-learning rag search-engine nlp large-language-models machine-learning python cross-platform embedding-models reranker bilingual crosslingual semantic-search sentence-transformers langchain llama-index retrieval-augmented-generation natural-language-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="netease-youdao/BCEmbedding")