Tabby

App in the BluixApps catalog

What it is

Tabby is a self-hosted AI coding assistant — a Copilot alternative that runs on your infrastructure. Code completion, chat-with-codebase, repository context, all via local LLM. Native IDE plugins for VS Code, JetBrains, Vim, Emacs. Designed for teams where proprietary code can't leave the network.

It's the answer to "I want Copilot but my legal department won't let me ship our codebase to GitHub".

What it's for

  • Code completion — Copilot-style inline suggestions on your code
  • Chat with codebase — ask questions about your repo via natural language
  • Code review assist — AI review on pull requests before human review
  • Refactoring helper — bulk transformations across files with AI
  • Privacy-bound development — proprietary code stays in-network

Who it's for

  • Enterprises with code privacy policies rejecting GitHub Copilot for compliance
  • Defense / government dev teams under no-cloud-code mandates
  • Financial services working on proprietary trading algos
  • Open-source organizations wanting self-hosted dev tooling
  • Indie developers preferring self-hosted to per-seat SaaS billing

Why teams pick Tabby over alternatives

  • Self-hosted Copilot — full feature parity with GitHub Copilot UX
  • Apache 2.0 — commercial use unrestricted
  • Code privacy — your code never leaves your network
  • Multiple IDE support — VS Code, JetBrains, Vim, Emacs
  • Repository context — RAG over your codebase, not just current file
  • Team management — multi-user with usage tracking

Integrations

  • IDE plugins — VS Code, JetBrains (IntelliJ, PyCharm, GoLand), Vim, Emacs
  • LLM backends — Tabby ships with optimized code models; supports custom OpenAI-compatible
  • Repository sources — Git, GitLab, GitHub, Gitea, Bitbucket
  • Authentication — built-in users + LDAP / OAuth via reverse proxy
  • Code search — semantic search over indexed repositories
  • Webhook — code review automation on PR events
  • API — REST endpoints for programmatic access

Notable users & community

  • 23k+ GitHub stars
  • Active Slack community
  • Featured in self-hosted dev tools guides
  • Backed by TabbyML with sustainable commercial enterprise offering
  • Continuous model improvements tracking SOTA code LLMs

Tips & operations

  • GPU strongly recommended — code completion latency on CPU is too high for usable IDE experience
  • Repository indexing is slow — first index of large monorepos takes hours; let it run
  • Update model frequently — code LLMs evolve fast; quarterly model swaps improve completion quality
  • Set up SSO — multi-user setup needs OAuth / LDAP from day one; avoid local user proliferation
  • Watch token costs if using OpenAI — code completion = high token volume; local model preferred
  • Persistent volume — indexed repos + chat history live on disk; mount from day one

What we ship in BluixApps

  • Docker compose: Tabby server + persistent index volume
  • Pinned tabbyml/tabby:latest (release-tagged)
  • HTTPS via Let's Encrypt
  • Default model: StarCoder2 for code completion
  • GPU passthrough optional (highly recommended for production)
  • Admin user with random password on first boot
  • Backup hook covers index + user data
Read this app's deep dive on bluix.app ↗

Get this app — pick a BluixApps plan

Same catalog. Scaling tenant isolation, white-label and support tier.

TierTenantsCatalogSupportWhite-labelMonthly
Stacks119 curated stacksStandard$19/moDetailDeploy
Starter10Full catalogStandard+$15–25/mo$49/moDetailDeploy
Pro25Full catalogPriority bugfix+$15–25/mo$149/moDetailDeploy
Growth100Full catalogPriority bugfix+$15–25/mo$349/moDetailDeploy
Scale500Full catalog7-day window+$15–25/mo$799/moDetailDeploy
EnterpriseUnlimitedFull catalogPriority 7-dayBundled$1,499/moDetailDeploy

Powered by WHMCompleteSolution