IntelLabs/fastRAG
An Intel research framework for building efficient retrieval-augmented generation pipelines using state-of-the-art LLMs and neural information retrieval.
Velocity · 7d
+1.4
★ / day
Trend
→steady
star history
fastRAG provides optimized RAG pipelines combining neural retrieval models like ColBERT with LLMs for generation. It supports semantic search, multi-modal data, knowledge graphs, and integrates with the Haystack ecosystem. The framework includes benchmarks, embedding optimizations, and tools for building advanced question-answering and summarization applications.