Siddhant-K-code/distill
Reliable LLM outputs start with clean context. Deterministic deduplication, compression, and caching for RAG pipelines.
Implements a deterministic context pipeline using agglomerative clustering, MaxMal Relevance re-ranking, and semantic deduplication without LLM calls—achieving ~12ms processing overhead. Supports multiple deployment modes: standalone API, vector database integration (Pinecone/Qdrant), and MCP protocol for Claude and AI assistants, with optional persistent memory featuring write-time deduplication and hierarchical decay for managing context across extended agent sessions.
136 stars.
Stars
136
Forks
14
Language
Go
License
AGPL-3.0
Last pushed
Feb 24, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/Siddhant-K-code/distill"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
louisbrulenaudet/ragoon
High level library for batched embeddings generation, blazingly-fast web-based RAG and quantized...
pesu-dev/ask-pesu
A RAG pipeline for question answering about PES University
namtroi/RAGBase
Open Source RAG ETL Platform. Turns PDFs, Docs & Slides into queryable vectors. Features a...
B-A-M-N/FlockParser
Distributed document RAG system with intelligent GPU/CPU orchestration. Auto-discovers...
aws-samples/rag-with-amazon-postgresql-using-pgvector-and-sagemaker
Question Answering application with Large Language Models (LLMs) and Amazon Postgresql using pgvector