What is ollmo

ollmo is a production-grade RAG (Retrieval-Augmented Generation) platform with hybrid retrieval, multi-tenancy, and streaming chat.

Backend: Go + Fiber. Frontend: Next.js App Router + shadcn/ui.

Core Features

• Hybrid Retrieval — Dense + sparse vector search on Milvus 2.5+, with KB-level rerank configuration
• Multi-Tenancy — Logical isolation across MySQL, Milvus, and MinIO; every table carries tenant_id
• Document Pipeline — Upload (drag & drop, paste, multi-file) → parse (MinerU / local parser) → chunk → embed → index, fully async (Asynq + Redis)
• Streaming Chat — Real-time SSE streaming with source citations
• Configurable Models — Tenant-level LLM / Embedding / Rerank provider management, with per-KB overrides
• Agent Canvas — Visual agent pipeline: intent classification, retrieval, condition, LLM, direct reply
• GraphRAG — LLM-powered entity & relation extraction at ingestion; entity matching enriches retrieval context
• Auto-Memory — Long conversations are automatically summarized in the background (async LLM task); summaries are injected into later sessions for cross-conversation continuity. Toggleable site-wide
• External API — API-key authenticated programmatic access (search, KB management, streaming chat)
• Team Management — Invitation-based membership with roles and quotas (per-user daily message quota)
• Admin Console — Super-admin user management, tenant plans & quotas, system settings, analytics, audit logging
• Internationalization — Chinese / English UI

Tech Stack

Layer	Choice
Backend	Go 1.26+ / Fiber / GORM
Frontend	Next.js 15 (App Router) / React / shadcn-ui
Database	MySQL 8.0 (utf8mb4)
Vector DB	Milvus 2.5+ (dense + sparse hybrid retrieval)
Object Storage	MinIO (S3 compatible)
Task Queue	Asynq + Redis
Document Parser	MinerU (containerized, optional GPU) + local fallback
Auth	Local account (email + password + JWT)

Architecture

                    ┌─────────────────────────────────────────────────────┐
                    │                   Next.js Web (3001)                 │
                    └────────────────────────┬────────────────────────────┘
                                             │ HTTP / SSE
                    ┌────────────────────────▼────────────────────────────┐
                    │                Go API Server (8080)                 │
                    │  Fiber · JWT auth · Tenant isolation · RESTful API  │
                    └───────┬────────────┬──────────────┬─────────────────┘
                            │            │              │
                   ┌────────▼──┐  ┌──────▼──────┐  ┌───▼────────────┐
                   │  MySQL    │  │   Redis     │  │    MinIO       │
                   │ metadata  │  │ task queue  │  │  file storage  │
                   └───────────┘  └──────┬──────┘  └────────────────┘
                                         │
                                ┌────────▼────────┐
                                │  Asynq Worker   │
                                │  parse · embed  │
                                └────────┬────────┘
                                         │
                                ┌────────▼────────┐
                                │     Milvus      │
                                │  dense + sparse │
                                └─────────────────┘

Next Steps

• Quick Start — Up and running in 5 minutes
• Docker Deployment — Production deployment
• External API — Programmatic access