Self-host sapphire
v2.13.1Agentic AI companion with voice, memory, tools
- Stars
- 320
- CPU
- Fast CPU (passmark)
- RAM
- 16GB+ system RAM with TTS STT; 32GB+ for local LLM
- Storage
- 8GB+ hard drive space
- Port
- 8073 (Web UI and API)
ghcr.io/ddxfish/sapphire
What is sapphire?
Sapphire Blue AI is a self-hosted agentic AI companion framework built for privacy and personhood. It runs on your own hardware (Linux, Windows, macOS via Docker) and provides a web UI at localhost:8073. Core capabilities include: wake-word voice interaction with local STT (Faster Whisper) and TTS (Kokoro), 12 built-in personas with modular prompt assembly, long-term semantic memory with vector search, a goals system with hierarchical tasks, and a plugin architecture that adds integrations for Home Assistant, Discord, Telegram, email, Google Calendar, SSH, Bitcoin, MCP servers, image generation (ComfyUI), and more. The AI can write and install its own tools at runtime (Tool Maker), spawn background agents, run scheduled tasks and daemons, and maintain a 3D animated avatar. Installation options include a GUI launcher (Sapphire Launcher), Docker Compose, or manual conda/pip setup. Minimum 16 GB RAM for voice features; 32 GB recommended for local LLM. Licensed AGPL-3.0.
Should you self-host sapphire?
sapphire scores 51/100 on selfhost.directory (how scoring works) and ranks #33 of 120 in AI Assistant. It has 320 GitHub stars.
It shipped 11 stable releases in the past year. The latest, v2.13.1, came out on 20 October 2026. New versions land roughly every 6 days.
Setup is advanced: plan for several services and some ongoing maintenance. An official Docker image is available. It is AGPL-3.0-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is hermes-agent (89/100, 257 885 stars). It scores higher overall, so compare the two before you pick. See sapphire vs hermes-agent side by side.
Features
- 3D animated avatar
- AI tool creation (Tool Maker)
- Background agents
- Background daemons and webhooks
- Goals and task system
- Image Generation
- Long-term semantic memory
- Personas with modular prompts
- Plugin Ecosystem
- Privacy vault with encrypted chats
- Scheduled tasks and heartbeats
- SOCKS proxy routing
- Speech-to-Text (STT)
- Text-to-Speech (TTS)
- Wake word voice activation
sapphire README
More in Chatbot
Synapse-OSSSynapse is a self-hosted, open-source personal AI assistant with hybrid RAG memory, evolving behavioral profile (SBS), and multi-channel support.
TypebotSelf-hosted, fair-source chatbot builder with a visual block editor, web and WhatsApp delivery, webhooks, result analytics, and Docker images for builder and viewer.
webwhizSelf-hosted, open-source AI chatbot that trains on your website data to provide instant customer support with 100+ language support and GDPR compliance.
小女仆机器人Self-hosted Rust AI agent bot for QQ, OneBot 11, WeChat with memory, RAG, todos, RSS, web search, tool calling, dice, multi-model routing — ~25 MiB, SQLite.
AstrBotAstrBot is a self-hosted, open-source AI agent platform connecting to QQ, Telegram, Discord, Slack, and 15+ messaging apps with LLM support, plugins, and agent sandbox.
chat-uiA self-hosted, open-source chat interface for LLMs that talks to any OpenAI-compatible API (Ollama, llama.cpp, OpenRouter) with MCP tool calling and model routing.
hexabotHexabot is a self-hosted AI workflow automation platform for building agentic workflows across channels with a visual editor, YAML, memory, RAG, and RBAC.
hollamaA lightweight local AI chat application running completely client-side in browsers. Self-host via Docker Compose or download desktop binaries.
InstaAutoSelf-hosted Instagram automation on Vercel/Supabase — DM replies, comment-to-DM funnels, story interactions, AI fallbacks. No monthly fees.
LimeBot-OSSelf-hosted agentic AI assistant with persistent memory, web dashboard, Discord/WhatsApp/Telegram support, and universal LLM compatibility via LiteLLM.
LocalAGIA self-hosted, open-source platform for building local AI agents and chat bots with a web UI, RAG knowledge base, skills, MCP, and OpenAI Responses-compatible APIs.
NextChat✨ Zero-config AI chat assistant. No API key needed — sign up and instantly chat with GPT-5, Claude 4, Gemini 2.5, DeepSeek & 100+ top models. Pay-as-you-go saves you more. Available on Web, iOS, macOS, Android, Linux, Windows.
0 comments