AI Index
Context management
Retrieval, compression, and context-routing products that decide what an agent sees. 27 entries: 16 commercial · 7 open source + paid cloud · 4 open source.
Amazon Web Services
Fully managed grounding service: multimodal parsing, embeddings, vector storage and retrieval for agents$5.00/GB raw data/mo index storage; $1.00/1k Retrieve API calls; agentic retr…
CommercialAzure AI Search (Foundry IQ)Microsoft
Azure search service with agentic retrieval pipeline and permission-aware knowledge bases for agentsDedicated hourly tiers plus serverless (preview); agentic retrieval billed pe…
CommercialCohere RerankCohere
Retrieval-optimization model API that reranks agent-retrieved documents before they reach the LLMOpen sourceCompresrOpen-source context-compression proxy (Context Gateway) for coding agents★ 647 on GitHub
CommercialContextual AIUnified context layer for enterprise RAG agents: datastores, query rewriting, rerank, grounded generationOn-demand PAYG; Parse $3/1k pages basic, $40/1k standard multimodal; Rerank-v…
Open source + paid cloudElastic Agent Builder (Elasticsearch)Elastic
Search-and-agent platform: hybrid retrieval, reranking, inference and agent skills over one index★ 78,549 on GitHub
CommercialExaNeural web search, page-content extraction and a hosted agent API for grounding on live web dataFree $10 credits/mo; pay-as-you-go from $4/1k search requests; agent effort t…
Open source + paid cloudFirecrawlWeb crawl and scrape API producing clean markdown pages for agent context windows★ 200,338 on GitHub
CommercialGleanEnterprise context and agents: 250+ connectors, permission-aware knowledge grounding, agent builderCommercialGoogle Agent Search (formerly Vertex AI Search)Google Cloud
Google-quality enterprise search and grounding service with out-of-the-box and DIY grounding APIsOpen sourceGraphRAGMicrosoft
Graph-augmented RAG retrieval infrastructure (adjacent, not memory product)★ 36,620 on GitHub · +6 today
Open source + paid cloudJina AI (Reader / Embeddings / Reranker)Elastic
Search-foundation APIs: URL-to-markdown reader, multimodal embeddings, listwise rerankers, OCR parser★ 12,251 on GitHub
CommercialLlamaCloud (LlamaParse / LlamaExtract / Index)LlamaIndex
Hosted document parsing, extraction and indexing platform for building document agentsFree 10K credits/mo; Starter $50/mo (40K credits, PAYG to $500/mo); Pro $500/…
Open sourceLMCacheKV-cache reuse layer (context infrastructure, not agent memory)★ 12,326 on GitHub · +3 today
Open source + paid cloudOnyxOpen Source AI Platform - AI Chat with advanced features that works with every LLM★ 33,321 on GitHub · +8 today
CommercialPineconeOpen sourceRAGFlowRAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.★ 93,949 on GitHub · +28 today
CommercialRagieManaged context engine: multimodal indexing, agentic OCR, entity extraction and retrieval APIsFree developer tier; Starter $100/mo; Pro $500/mo; extra pages from $0.02; st…
CommercialRedis IrisRedis
CommercialTavilySearch, extract and crawl API marketed specifically as the context feed for AI agentsFree 1,000 API credits/mo; pay-as-you-go $0.008/credit; Project $30/mo for 4,…
CommercialturbopufferServerless vector and full-text search database built on object storage, with namespaces per tenant$16/mo minimum usage (Launch), $256/mo (Scale), enterprise from $4,096/mo; st…
Open source + paid cloudUnstructuredDocument parsing and structuring platform turning messy files into agent-ready element-level JSON★ 15,709 on GitHub
CommercialVectaraEnterprise platform for grounded AI agents: corpus ingestion, retrieval LLM, citations and governance30-day free trial; SaaS from $100K/yr, VPC from $250K/yr, on-prem from $500K/yr
CommercialVertex AI RAG Engine (Gemini Enterprise Agent Platform)Google Cloud
Managed RAG framework for context augmentation: ingestion, chunking, embeddings, retrieval, rerankingBilled per component: embedding model usage, Spanner/vector database for inde…
CommercialVoyage AIMongoDB (Voyage AI)
Embedding and reranker model API tuned for retrieval quality on unstructured and multimodal datavoyage-4 $0.06/1M tokens, voyage-4-large $0.12, voyage-4-lite $0.02, voyage-c…
Open source + paid cloudWeaviate CloudWeaviate
Managed vector database with hybrid search, hosted embeddings and a natural-language Query Agent★ 16,945 on GitHub
Open source + paid cloudZilliz Cloud (managed Milvus)Zilliz
Fully managed vector database service built on Milvus, with serverless and dedicated cluster tiers★ 46,629 on GitHub