Build, deploy and orchestrate AI agents, RAG pipelines, workflows, and more. One platform, every AI capability.
NexusAI combines the best ideas from 11 open-source AI projects into one self-hosted platform:
| Feature | Inspired By |
|---|---|
| Multi-agent orchestration | CrewAI, AutoGen |
| Graph-based stateful agents | LangGraph |
| AI developer agent | OpenManus, Devin |
| Visual workflow builder | n8n, Langflow |
| RAG pipelines | Dify, Flowise |
| Audio transcription | Whisper |
| Backend platform | Supabase |
| LLM gateway | OpenAI proxy |
git clone https://github.com/YOUR_USERNAME/nexusai.git
cd nexusai
cp .env.example .env
# Edit .env — add your NVIDIA_API_KEY at minimumdocker compose up -dhttp://localhost:3000 # NexusAI UI
http://localhost:8000/docs # API docs (Swagger)
http://localhost:3001 # Grafana monitoring
http://localhost:9090 # Prometheus metrics
┌─────────────────────────────────────────┐
│ NexusAI Dashboard │
│ Next.js 14 + React Flow │
├─────────────┬───────────────────────────┤
│ Agents │ Workflows │ Chat │RAG │
├─────────────┴───────────────────────────┤
│ FastAPI Backend (Python) │
│ Multi-Agent Core │ Dev Agent │ Audio │
├─────────────────────────────────────────┤
│ PostgreSQL + pgvector │ Redis │ Celery │
├─────────────────────────────────────────┤
│ NVIDIA │ OpenAI │ Groq │ Mistral │ ... │
└─────────────────────────────────────────┘
All config lives in .env. Key variables:
NVIDIA_API_KEY=nvapi-... # Your NVIDIA API key
NVIDIA_DEFAULT_MODEL=openai/gpt-oss-120b
# Optional additional providers
OPENAI_API_KEY=sk-...
GROQ_API_KEY=gsk_...
MISTRAL_API_KEY=...cd backend
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -r requirements.txt
uvicorn app.main:app --reloadcd frontend
npm install
npm run dev# Backend
cd backend && pytest tests/ -v
# Load test
cd backend && locust -f tests/load_test.py --host=http://localhost:8000
# Frontend E2E
cd frontend && npx playwright testFull Swagger docs at http://localhost:8000/docs
Key endpoints:
| Endpoint | Method | Description |
|---|---|---|
/api/v1/agents/ |
GET/POST | List / create agents |
/api/v1/agents/{id}/run |
POST | Run an agent |
/api/v1/agents/crew/run |
POST | Run a multi-agent crew |
/api/v1/agents/dev/run |
POST | Run AI developer agent |
/api/v1/workflows/ |
GET/POST | Workflows CRUD |
/api/v1/rag/kb |
POST | Create knowledge base |
/api/v1/rag/kb/{id}/upload |
POST | Upload document |
/api/v1/rag/kb/{id}/query |
POST | RAG query |
/api/v1/llm/chat |
POST | Direct LLM chat |
/api/v1/audio/transcribe |
POST | Transcribe audio |
/health |
GET | Health check |
/metrics |
GET | Prometheus metrics |
See CONTRIBUTING.md. Issues and PRs welcome!
MIT