mirror of
https://github.com/NVIDIA/dgx-spark-playbooks.git
synced 2026-04-24 10:53:52 +00:00
8974ee9913
- Add query mode badge to answer section showing Pure RAG/Traditional Graph/GraphRAG - Add collapsible reasoning section for <think> tags in answers - Add markdown rendering support (bold/italic) in answers - Fix Pure RAG to properly display answers using llmAnswer state - Hide empty results message for Pure RAG mode - Update metrics sidebar to show query times by mode instead of overall average - Add queryTimesByMode field to metrics API and frontend interfaces - Disable GraphRAG button with "COMING SOON" badge (requires GNN model) - Fix Qdrant vector store document mapping with contentPayloadKey - Update console logs to reflect Qdrant instead of Pinecone - Add @qdrant/js-client-rest dependency to package.json 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| app | ||
| components | ||
| contexts | ||
| hooks | ||
| lib | ||
| public | ||
| styles | ||
| types | ||
| utils | ||
| components.json | ||
| next.config.mjs | ||
| next.env.d.ts | ||
| package.json | ||
| postcss.config.mjs | ||
| README.md | ||
| tailwind.config.ts | ||
| tsconfig.json | ||
Frontend Application
This directory contains the Next.js frontend application for the txt2kg project.
Structure
- app/: Next.js 15 app directory with pages and API routes
- API routes for LLM providers (Ollama, vLLM, NVIDIA API)
- Triple extraction and graph query endpoints
- Settings and health check endpoints
- components/: React 19 components
- Graph visualization (Three.js WebGPU)
- PyGraphistry integration for GPU-accelerated rendering
- RAG query interface
- Document upload and processing
- contexts/: React context providers for state management
- hooks/: Custom React hooks
- lib/: Utility functions and shared logic
- LLM service (Ollama, vLLM, NVIDIA API integration)
- Graph database services (ArangoDB, Neo4j)
- Pinecone vector database integration
- RAG service for knowledge graph querying
- public/: Static assets
- types/: TypeScript type definitions for graph data structures
Technology Stack
- Next.js 15: React framework with App Router
- React 19: Latest React with improved concurrent features
- TypeScript: Type-safe development
- Tailwind CSS: Utility-first styling
- Three.js: WebGL/WebGPU 3D graph visualization
- D3.js: Data-driven visualizations
- LangChain: LLM orchestration and chaining
Development
To start the development server:
cd frontend
npm install
npm run dev
Or use the start script from project root:
./start.sh --dev-frontend
The development server will run on http://localhost:3000
Building for Production
cd frontend
npm run build
npm start
Or use Docker (recommended):
# From project root
./start.sh
The production app will run on http://localhost:3001
Environment Variables
Required environment variables are configured in docker-compose files:
ARANGODB_URL: ArangoDB connection URLOLLAMA_BASE_URL: Ollama API endpointVLLM_BASE_URL: vLLM API endpoint (optional)NVIDIA_API_KEY: NVIDIA API key (optional)PINECONE_HOST: Local Pinecone host (optional)SENTENCE_TRANSFORMER_URL: Embeddings service URL (optional)
Features
- Knowledge Graph Extraction: Extract triples from text using LLMs
- Graph Visualization: Interactive 3D visualization with Three.js WebGPU
- RAG Queries: Query knowledge graphs with retrieval-augmented generation
- Multiple LLM Providers: Support for Ollama, vLLM, and NVIDIA API
- GPU-Accelerated Rendering: Optional PyGraphistry integration for large graphs
- Vector Search: Pinecone integration for semantic search