Self-host orbit
v2.21.0AI gateway for private RAG and tool-calling agents
- Stars
- 357
- RAM
- 4GB+
- Storage
- 3GB+
- Ports
- 3000 (HTTP API server), 5173 (OrbitChat web UI), 3443 (HTTPS API server (optional))
schmitech/orbit-ollama
What is orbit?
ORBIT (Open Retrieval-Based Inference Toolkit) is a self-hosted AI gateway that connects private data — files, SQL databases, NoSQL stores, vector databases, APIs, and MCP tools — to local or hosted language models through one OpenAI-compatible API. It supports multiple inference providers including Ollama (local), OpenAI, Google Gemini, OpenRouter, Cohere, Azure OpenAI, Anthropic, vLLM, llama.cpp, and SGLang. The platform includes built-in authentication with username/password, optional OIDC/SSO via Microsoft Entra ID and Auth0, two-factor authentication (TOTP), API key management, quotas, audit logging, rate limiting, and a web-based admin dashboard. ORBIT can run on your own infrastructure or in a private cloud with local models for air-gapped deployments. It provides a multimodal chat interface (OrbitChat) for document Q&A, voice conversations, and tool-calling agents. Deployment options include pre-built Docker flavor images (Ollama, OpenAI, Gemini, OpenRouter, Cohere), Docker Compose with separate Ollama container, binary tarball releases, and local Python installation.
Should you self-host orbit?
orbit scores 51/100 on selfhost.directory (how scoring works) and ranks #44 of 125 in AI. It has 357 GitHub stars.
It shipped 18 stable releases in the past year. The latest, v2.21.0, came out on 28 October 2026. New versions land roughly every 2 days.
Setup is moderate: expect some configuration beyond a single command. An official Docker image is available. It is Apache-2.0-licensed and written mostly in Python.
Its closest alternative on selfhost.directory is ComfyUI (93/100, 143 668 stars). It scores higher overall, so compare the two before you pick. See orbit vs ComfyUI side by side.
Features
- Admin Dashboard
- Air-gapped deployment support
- Audit Logging
- Automatic model selection
- Built-in authentication and RBAC
- Decision model adapters
- File processing and vector storage
- MCP (Model Context Protocol) tool calling
- Multimodal chat (OrbitChat)
- Multi-Model Support
- OIDC/SSO integration
- OpenAI Compatible API
- RAG (Retrieval-Augmented Generation)
- Rate limiting and throttling
- Two-factor authentication (TOTP)
orbit README
More in RAG
paperless-iqPaperless IQ is a self-hosted AI add-on for Paperless-NGX adding metadata tagging, RAG search, vision analysis, and approval workflow — supports Ollama, Anthropic, OpenAI
PharosSelf-hosted, open-source RAG system with multi-format document ingestion, hybrid retrieval, enterprise ACL, and dual HTTP + MCP exits for team knowledge bases.
private-gptOpen-source Claude-compatible API layer adding RAG with citations, tools, MCP and SQL queries to any local model server, plus a self-hosted workbench UI.
qurioQurio is a self-hosted, open-source RAG engine that ingests docs and code locally, then serves grounded context to AI coding assistants via MCP to prevent hallucinations.
R2RSelf-hosted agentic RAG system with a RESTful API: multimodal ingestion, hybrid search, knowledge graphs, a Deep Research agent, and user & access management.
ragSelf-hosted, open-source retrieval-augmented generation chat app on txtai and Streamlit: vector RAG plus graph path queries, with your own LLM.
SeekStormOpen-source Rust search engine with BM25 lexical and vector search fused in one hybrid query, multi-tenant REST server, web UI and MCP support. Self-hosted, free forever.
supermemorySelf-hosted AI memory and context server: Supermemory Local extracts facts, builds user profiles, and blends RAG with memory search through one API on port 6767.
surfsenseSelf-hosted, open-source NotebookLM alternative: an air-gapped desktop AI agent that researches, edits and transforms your documents with any local or hosted model.
TaskingAIAn open-source, self-hosted BaaS stack for building LLM agents: unified model APIs, RAG, built-in tools and an admin console, deployed with Docker Compose.
UtopiaSelf-hosted, open-source enterprise world model: a Rust binary plus Postgres/pgvector holding a bitemporal knowledge graph, with agentic RAG chat and an MCP server.
VaneSelf-hosted, open-source AI answering engine: searches via bundled SearxNG and answers with cited sources, using local Ollama models or OpenAI, Claude and Gemini.
0 comments