Neura Market
GitHub logo

GitHub

Free

The context layer for AI agents

FreeFree tier
Type
Open Source
Company
Graphlit
Visit Website GitHub (40 stars)

About GitHub

Graphlit is a cloud-native platform that gives AI applications semantic memory. It goes beyond vector search to provide real knowledge retrieval with context, relationships, and understanding. Graphlit offers a single API for the complete stack: content ingestion, extraction, enrichment, storage, and retrieval. It automatically extracts entities, relationships, and summaries from ingested content, supports true multimodal data (documents, audio, video, images), and provides smart retrieval via semantic search, knowledge graph queries, and RAG-powered conversations. The platform integrates with best-in-class LLMs (OpenAI, Anthropic, Google, xAI, Deepseek, Groq, Mistral, Cohere, Cerebras, AWS Bedrock) and is MCP-native, connecting to tools like Cursor, VS Code, Windsurf, Claude Desktop, Claude Code, and ChatGPT. Official SDKs are available for TypeScript/JavaScript, Python, and C#/.NET. Graphlit is designed to save developers weeks of engineering time by eliminating the need to build and maintain complex ingestion-to-retrieval infrastructure.

Key Features

Ingest anything: documents (PDF, DOCX, PPTX, Excel, Markdown), media (audio transcription, video processing, image analysis), web scraping, RSS feeds, sitemaps, platforms (Slack, Gmail, Notion, GitHub, Jira, Linear, SharePoint), cloud storage (S3, Azure Blob, Google Drive, Dropbox, OneDrive, Box)
Automatic extraction: entity recognition and linking, relationship mapping, OCR and visual object detection, audio transcription with speaker diarization, automated summarization
Smart retrieval: semantic search (vector + hybrid), knowledge graph queries, RAG-powered conversations, multi-tenant filtering, context-aware results
Best-in-class LLM support: OpenAI, Anthropic, Google, xAI, Deepseek, Groq, Mistral, Cohere, Cerebras, AWS Bedrock (all support tool calling, streaming, and reasoning modes)
MCP-native integration: connect to Cursor, VS Code, Windsurf, Claude Desktop, Claude Code, ChatGPT via npx graphlit-mcp-server
Official SDKs: TypeScript/JavaScript (npm install graphlit-client), Python (pip install graphlit-client), C# / .NET (dotnet add package Graphlit.Client)

Pros & Cons

Pros
  • One API for complete ingestion-to-retrieval stack, saving weeks of engineering time
  • Cloud-native platform with no infrastructure to manage
  • Years of production hardening and true multimodal support from day one
  • Free tier includes 1GB storage, 1K content items, 3 feeds, 100 conversations, all content types, full API access, and community support
  • No credit card required to start building
Cons
  • Limited free tier capacity (1GB storage, 1K content items) may not suit large-scale projects
  • Dependency on cloud services; not available as self-hosted option
  • Less fine-grained control over custom retrieval pipelines compared to building from scratch

Best For

AI Agents Copilots – give your AI memory and contextKnowledge Management – build searchable repositories from unstructured dataDocument Intelligence – extract insights from PDFs, reports, contractsCustomer Support – RAG-powered chatbots over your documentationResearch Tools – semantic search across academic papers, articlesMedia Analysis – transcribe and analyze audio/video content

FAQ

What is Graphlit?
Graphlit is a cloud-native semantic memory platform for AI agents. It provides a single API for ingesting, extracting, enriching, storing, and retrieving content with context, relationships, and understanding.
How do I get started with Graphlit?
You can start for free in 5 minutes with no credit card required. The free tier includes 1GB storage, 1K content items, 3 feeds, 100 conversations, and full API access.
What LLMs does Graphlit support?
Graphlit supports OpenAI, Anthropic (Claude), Google (Gemini), xAI (Grok), Deepseek, Groq, Mistral, Cohere, Cerebras, and AWS Bedrock. All models support tool calling, streaming, and reasoning modes.
Does Graphlit support multimodal content?
Yes, Graphlit is truly multimodal from day one. It supports documents (PDF, DOCX, PPTX, Excel, Markdown), audio (transcription with speaker diarization), video processing, and image analysis (OCR, visual object detection).