Firecrawl
Turns websites into LLM-ready Markdown or structured data for agents.
Website captured Oct 19
Retrieve
Pull the right outside knowledge into context, just in time: search, RAG, embeddings, rerankers, docs and code fetchers.
30 listed entries
tool
MarkItDown
Converts files and Office documents to Markdown, the most token-friendly format for a context window.
RAGFlow
Open-source RAG engine with deep document understanding and built-in agent workflows.
Website captured Oct 19
Crawl4AI
Self-hostable crawler that turns any website into clean, LLM-ready Markdown.
Website captured Oct 19
Docling
Parses PDFs, Office files and scans into clean, structured documents ready for chunking and retrieval.
Website captured Oct 19
Context7
Pulls up-to-date, version-specific library docs and code examples into a coding agent's prompt, via MCP.
LlamaIndex
Document parsing, indexing and retrieval framework that turns your data into context an agent can query.
Website captured Oct 19
Milvus
Cloud-native vector database built for large-scale approximate nearest-neighbor search.
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
LightRAG
Graph-plus-vector RAG that retrieves entities and relations, not just chunks (EMNLP 2025).
Website captured Oct 19
GraphRAG
Microsoft's graph-based RAG: builds a knowledge graph and community summaries to answer whole-corpus questions.
Website captured Oct 19
Qdrant
Open-source vector database and search engine for semantic and hybrid retrieval at scale.
Website captured Oct 19
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
Chroma
Open-source search infrastructure for AI: embed, store and query documents for retrieval.
Website captured Oct 19
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
pgvector
Vector similarity search inside Postgres, so retrieval lives next to your app data.
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
Sentence Transformers
The standard Python library for embedding, retrieval and reranking models.
Website captured Oct 19
Weaviate
Open-source vector database that combines vector search with structured filtering.
Website captured Oct 19
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
Unstructured
Open-source ETL that turns messy documents into clean, structured elements for RAG pipelines.
Website captured Oct 19
txtai
All-in-one framework for semantic search, embeddings databases and LLM workflows.
Website captured Oct 19
LanceDB
Embedded retrieval library for multimodal data: vector, full-text and hybrid search inside your process.
Website captured Oct 19
Disclosure: the operator of shipping.buzz is a Qdrant Star Ambassador. No manual overrides apply to vector databases.
Chonkie
Lightweight, fast chunking library (Python and TypeScript) for splitting text before embedding.
Website captured Oct 19
BGE-M3
Multilingual embedding model that does dense, sparse and multi-vector retrieval in one model (BAAI).
Website captured Oct 19
Contextual retrieval
Prepend chunk-specific context before embedding and BM25 indexing. Anthropic measured 49% fewer failed retrievals, and 67% fewer with reranking.
EmbeddingGemma 300M
Google's 300M-parameter embedding model, small enough to run on device.
Website captured Oct 19
Exa
Neural web search API for agents: finds pages by meaning and returns their contents.
Website captured Oct 19
Harrier OSS v1 0.6B
Microsoft's open 0.6B embedding model (MIT) for retrieval and semantic search.
Website captured Oct 19
Just-in-time retrieval
Keep lightweight references (file paths, stored queries, links) in context and load the data with tools only when it's needed.
mxbai-rerank-large-v2
Open reranking model from Mixedbread that reorders search results before they reach the prompt.
Website captured Oct 19
Qwen3-Embedding-0.6B
Small multilingual text-embedding model from the Qwen3 family.
Website captured Oct 19
Qwen3-Reranker-0.6B
Small Qwen3 reranker: reorders retrieved passages so only the best reach the prompt.
Website captured Oct 19
Rerank before you fill the prompt
Retrieve broadly (vector plus keyword), then rerank and pass only the top few chunks to the model.
Tavily
Search API built for agents: returns ranked, extracted web results ready for the context window.
Website captured Oct 19
How they connect
- Docling alternative to Unstructured
- Docling alternative to MarkItDown
- Crawl4AI alternative to Firecrawl
- Chroma alternative to Qdrant
- Chroma alternative to pgvector
- Chroma alternative to LanceDB
- BGE-M3 alternative to Qwen3-Embedding-0.6B
- BGE-M3 alternative to Harrier OSS v1 0.6B
- BGE-M3 alternative to EmbeddingGemma 300M
- Rerank before you fill the prompt implements mxbai-rerank-large-v2
- Rerank before you fill the prompt implements Qdrant
- Rerank before you fill the prompt implements Weaviate
- Rerank before you fill the prompt implements Qwen3-Reranker-0.6B
- Qwen3-Embedding-0.6B alternative to EmbeddingGemma 300M
- GraphRAG alternative to LightRAG
- Harrier OSS v1 0.6B alternative to Qwen3-Embedding-0.6B
- Exa alternative to Tavily
- Firecrawl alternative to Exa
- Firecrawl alternative to Tavily
- Unstructured alternative to MarkItDown
- mxbai-rerank-large-v2 alternative to Qwen3-Reranker-0.6B
- RAGFlow alternative to LlamaIndex
- Qdrant alternative to Milvus
- Qdrant alternative to pgvector
- Qdrant alternative to Weaviate
- Contextual retrieval implements Sentence Transformers
- Contextual retrieval implements mxbai-rerank-large-v2
- Contextual retrieval implements Qwen3-Reranker-0.6B
- Contextual retrieval implements LlamaIndex
- Milvus alternative to Weaviate
- LanceDB alternative to Qdrant
- Just-in-time retrieval implements Context7
- LlamaIndex alternative to txtai