TokenPack packs long documents, codebases, PDFs, and folders into compact, evidence-dense LLM context using local embeddings, evidence scoring, and budget-aware selection.
python cli mcp agents semantic-search rag pdf-processing sentence-transformers document-ai llm long-context retrieval-augmented-generation context-compression llmlingua context-packing
-
Updated
Oct 16, 2026 - Python