Pinned Loading
-
agentic-rag
agentic-rag PublicA local, agentic RAG platform (Ollama + Qdrant + Postgres + Redis + MinIO) that runs via Docker Compose or Kubernetes , planner, hybrid retrieval, critic, and generator loop with full traceability.
Python
-
pocket-tts
pocket-tts PublicFully local voice agent: LangGraph + Ollama/OpenAI + Kyutai Pocket TTS (CPU-only, no GPU). Dockerized, with a Streamlit chat UI and streaming speech output.
Python
-
qubo-rag-knapsack
qubo-rag-knapsack PublicA research project exploring Quantum computing-inspired optimization for RAG, replacing naive Top-K retrieval with a QUBO-based knapsack optimizer that selects the most relevant combination of chun…
Python
-
rag-laya-poc
rag-laya-poc PublicRAG Showdown: Comparing naive vector search vs Laya-reranked retrieval in a dockerized RAG pipeline with OpenAI generation.
HTML
-
train-jev-model
train-jev-model PublicLightweight decision model built on Qwen2.5-0.5B with LoRA for calibrated multi-option decision-making, confidence scoring, and selective abstention.
Jupyter Notebook
If the problem persists, check the GitHub status page or contact support.

