slvDev
Featured
esp32-ai
A 28.9 million parameter language model runs on an $8 ESP32-S3 microcontroller entirely on-device, generating simple stories at about 9.5 tokens per second.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It combines dense and sparse retrieval with weighted RRF fusion, optional reranker, and includes rigorous evaluation metrics (Hit Rate, MRR, nDCG), ablation studies, and medical safety boundaries, all documented transparently.
Who it is for
- Developers building medical RAG applications
- Researchers in information retrieval for healthcare
- Ophthalmology educators or content creators
- AI engineers interested in reproducible evaluation
Use cases
- Answering eye health questions based on curated documents
- Comparing retrieval strategies (dense vs hybrid vs rerank) via built-in evaluation
- Learning RAG system architecture with emphasis on safety and evidence
- Prototyping a medical Q&A system that refuses low-relevance queries
Strengths
- Supports both vector (DashScope text-embedding-v4) and keyword (BM25) retrieval with weighted RRF fusion
- Optional gte-rerank-v2 reranker with automatic fallback
- Includes low-relevance refusal, safety warnings for urgent symptoms, and citation format
- Provides reproducible evaluation with multiple metrics and chunking ablation experiments
Considerations
- Evaluation set is very small (8 queries), limiting statistical significance
- Documents are short and not representative of real multi-chapter medical texts
- No faithfulness or answer-citation accuracy evaluation yet implemented
README quick start
安装与启动
Description
Ophthalmology RAG system with hybrid retrieval, rerank, citations and evaluation
Related repositories
Similar projects matched by category, topics, and programming language.
gavamedia
Featured
deltafin
Deltafin is a research project that runs the 2.8-trillion-parameter Mixture-of-Experts model Kimi K3 on a single Apple Silicon Mac (e.g., M1 Max with 64 GB) at about 16 seconds per token, using exact, reproducible inference with local or streaming expert loading.
jamesob
Featured
local-llm
A comprehensive guide for building and configuring a high-end local machine to run state-of-the-art LLMs, with detailed hardware choices, BIOS tuning, and Docker-based model serving.