Toolkit

We provides some basic tools to help you build the RAG pipeline. But it’s not limited to these internal tools. You can use whatever you like.

You may need to install with extras:

pip install vechord[openai,spacy,cohere]

• Augment

  • GeminiAugmenter: for contextual retrieval

• Chunk

  • RegexChunker: Regex based chunker

  • SpacyChunker: Spacy based chunker

  • GeminiChunker: Gemini based chunker

• Embedding

  • GeminiDenseEmbedding: Gemini embedding

  • OpenAIDenseEmbedding: OpenAI embedding

  • JinaDenseEmbedding: JinaAI embedding

  • VoyageDenseEmbedding: VoyageAI embedding

  • SpacyDenseEmbedding: Spacy embedding

• Evaluate

  • GeminiEvaluator: Gemini based query generator

  • GeminiUMBRELAEvaluator: Gemini UMBRELA evaluator

• Extract

  • SimpleExtractor: Simple extractor

  • GeminiExtractor: Gemini extractor

  • LlamaParseExtractor: Llama extractor

• Graph

  • SpacyEntityRecognizer: Spacy based entity recognizer

  • GeminiEntityRecognizer: Gemini based entity recognizer

• GroundTruth

  • GroundTruth: generate ground truth

• Rerank

  • CohereReranker: Cohere reranker

  • JinaReranker: Jina MultiModal reranker

  • ReciprocalRankFusion: fuse function for hybrid retrieval