Awesome-RAG
😎 Awesome list of Retrieval-Augmented Generation (RAG) applications in Generative AI.
https://github.com/Danielskry/Awesome-RAG
Last synced: 2 days ago
JSON representation
-
💾 Databases
-
Benchmarks
-
Other Database Systems:
- Neo4j
- Qdrant - source vector database designed for similarity search.
- Redis Stack - memory data structure store used as a database, cache, and message broker.
- Azure Cosmos DB - model database service with integrated vector search.
- Couchbase
- Lantern - aware personal search engine.
- LlamaIndex - memory vector store for rapid experimentation.
- SurrealDB - model database optimized for time-series data.
- Weaviate - source cloud-native vector search engine.
-
Distributed Data Processing and Serving Engines:
- Apache Cassandra
- MongoDB Atlas - model database service with integrated vector search.
- Vespa - source big data processing and serving engine designed for real-time applications.
-
Search Engines with Vector Capabilities:
-
Vector Databases:
- Chroma DB - native open-source embedding database.
- Milvus - source vector database for AI-powered applications.
- Pinecone
- Oracle AI Vector Search
-
Relational Database Extensions:
- Pgvector - source extension for vector similarity search in PostgreSQL.
-
Vector Search Libraries and Tools:
- FAISS - scale datasets and optimized for fast retrieval of nearest neighbors.
-
-
🧰 Frameworks that Facilitate RAG
- Swiftide
- Haystack - ready LLM applications.
- LangChain - purpose framework for working with LLMs.
- Semantic Kernel
- Verba - source application for RAG out of the box.
- Mastra
- Dify - source LLM app development platform.
- Cognita - source RAG framework for building modular and production ready applications.
- Letta
- Flowise
- LlamaIndex
- CocoIndex
-
🛠️ Techniques
-
Response quality & safety
- Hallucination
- Guardrails - checking.
- Prompt Injection Prevention
-
Prompting
-
Chunking
-
Embeddings
-
Retrieval
-
Data cleaning
- Data cleaning techniques - processing steps to refine input data and improve model performance.
-
-
🎯 Approaches
- RAG Fusion
- Temporal Augmented Retrieval - sensitive data in retrieval processes.
- Plan-then-RAG
- GraphRAG
- FLARE - An approach that incorporates active retrieval-augmented generation to improve response quality.
- Contextual Retrieval - Improves retrieval by adding relevant context to document chunks before retrieval, enhancing the relevance of information retrieved from large knowledge bases.
- Self Reflective RAG
- Agentic RAG
- GNN-RAG
- Corrective RAG
- Cache-Augmented Generation (CAG) - Value (KV) cache).
- Retrieval-Augmented Fine-Tuning - tune LLMs specifically for enhanced retrieval and generation tasks.
-
📊 Metrics
-
Search metrics
-
Response Evaluation Metrics
- LangFuse - source tool for tracking LLM metrics, observability, and prompt management.
- LangSmith - grade LLM applications, allows you to closely monitor and evaluate your application.
- Hugging Face Evaluate
- Weights & Biases
- Ragas
- BLEU - grams between machine-generated and reference outputs, providing insight into precision.
- ROUGE - grams, skip-bigrams, or longest common subsequence with reference outputs.
- METEOR
- TuringBench
- Hugging Face Evaluate
-
Categories
Sub Categories
Chunking
12
Response Evaluation Metrics
10
Other Database Systems:
9
Prompting
9
Search metrics
5
Retrieval
5
Vector Databases:
4
Distributed Data Processing and Serving Engines:
3
Response quality & safety
3
Search Engines with Vector Capabilities:
2
Embeddings
1
Data cleaning
1
Vector Search Libraries and Tools:
1
Benchmarks
1
Relational Database Extensions:
1
Keywords
llm
10
ai
8
rag
8
llmops
4
openai
4
python
4
semantic-search
3
typescript
3
nearest-neighbor-search
3
large-language-models
3
machine-learning
3
image-search
2
nextjs
2
hnsw
2
artificial-intelligence
2
workflows
2
graph-database
2
javascript
2
deep-learning
2
vector-database
2
vector-search
2
nosql
2
analytics
2
backend-as-a-service
2
evaluation
2
langchain
2
mlops
2
approximate-nearest-neighbor-search
2
data
2
indexing
2
gpt
2
gpt-4
2
llms
2
agents
2
distributed
2
chatgpt
2
generative-ai
2
information-retrieval
2
language-model
2
search-engine
2
pytorch
2
database
2
retrieval-augmented-generation
2
chatbot
2
cypher
1
ycombinator
1
graph
1
self-hosted
1
graphdb
1
neo4j
1