Built for
knowledge at scale
From document ingestion to AI-powered answers, Cortex provides everything you need to build intelligent knowledge systems.
Knowledge Graph & GraphRAG
Turn documents into connected knowledge
Cortex uses GraphRAG to automatically extract entities, relationships, and concepts from your documents. The result is a traversable knowledge graph that captures not just content, but meaning.
- Automatic entity extraction (people, orgs, concepts, locations)
- Relationship discovery between entities
- Semantic entity resolution (merge duplicates)
- Multi-hop graph traversal for context retrieval
- Community detection and summarization
Key Metrics
Hybrid Search
Three search methods, unified
Cortex combines vector similarity, keyword matching, and graph traversal using Reciprocal Rank Fusion. A cross-encoder re-ranks results for maximum precision.
- Vector search using your embeddings of choice
- Full-text keyword search with phrase matching
- Graph traversal finds related entities
- RRF fusion combines all methods
- Cross-encoder re-ranking for top results
Key Metrics
RAG Q&A with Streaming
Natural language, instant answers
Ask questions about your knowledge base and get accurate, cited answers. Cortex supports streaming responses, agentic multi-step reasoning, and conversation history.
- Natural language question answering
- Source citations with every answer
- Streaming responses for real-time UX
- Agentic mode with multi-step reasoning
- Conversation history for context
Key Metrics
Enterprise Security
Built for production
Cortex includes built-in protection against prompt injection attacks, granular API key permissions, and secure session management. Deploy with confidence.
- Prompt injection protection built-in
- API key authentication with roles
- Granular permissions (read, write, delete, admin)
- Key expiration and rotation
- Secure session management
Key Metrics
Collections & Organization
Structure at scale
Organize documents into collections for different teams, projects, or use cases. Each collection has its own scoped knowledge graph and search index.
- Assign documents to collections
- Scoped search within collections
- Collection-specific knowledge graphs
- Bulk document operations
Key Metrics
API-First Architecture
Build anything on Cortex
Cortex exposes 60+ REST API endpoints with full OpenAPI documentation. Integrate with your existing tools, build custom interfaces, or power AI agents.
- 60+ REST API endpoints
- Full OpenAPI/Swagger documentation
- Webhook support for events
- Background task tracking
- Comprehensive error handling
Key Metrics
Turbo Mode
Powered by comput3.ai
Launch dedicated GPUs with per-second billing through our partners at comput3.ai. Perfect for initial knowledge graph generation when uploading massive document sets—pay only for the compute you use.
- Per-second GPU billing via comput3.ai
- Dedicated B200 GPUs on demand
- Accelerate initial knowledge graph generation
- Process large document uploads faster
- Automatic fallback to standard mode
Key Metrics
Usage Analytics
Insights into your knowledge base
Track how your knowledge base is being used. Monitor query patterns, document access, and system performance to optimize your setup.
- Query volume and patterns
- Document access statistics
- Performance monitoring
- Usage reports and exports
- API call tracking
Key Metrics
Ready to explore?
See Cortex in action with our live demo or join the waitlist for early access.