Self-hosted coding agent server with team SSO, audit logs, and dashboards. Enterprise-grade.
Editorial verdict: “Best self-hosted server for teams. SSO + audit logs make it the IT-friendly pick.”
Which runtime + OS combos this app works against. Source of truth for "will it run on my setup?"
Tabby is the coding agent server you pick when you’re not just running local models for yourself, but deploying them to a team of 20 or more. It bridges llama-cpp or any OpenAI-compatible runtime into a single server that handles autocomplete, chat, and model serving, with editor extensions for VS Code, JetBrains, Vim, and Emacs. The real differentiator is enterprise plumbing: SSO, per-user audit logs, and dashboards that let you prove exactly which completions were generated by whom. You’ll want at least 16 GB VRAM and a DeepSeek Coder 6.7B Q4_K_M for fill-in-the-middle, plus a larger model for chat. The tradeoff is that this is more moving parts than a solo Continue setup, and the dashboard features live behind a paid tier.
Editor-integrated or CLI agent that edits code via your model.
Best terminal-native coding agent for local models. Qwen 2.5 Coder 32B is its sweet spot.
Best minimal-surface Copilot-replacement that's been Ollama-native since day one.
Best IDE-integrated agent that fully respects 'all local' as a first-class option.
Best Copilot replacement that defaults to local. Configurable; pair with Qwen 2.5 Coder.
Pre-filled with this app's recommended use case + budget tier. Get the full rig + runtime + model picks.
The full directory — filter by category, runtime, OS, privacy posture, or VRAM.
What this app talks to: Ollama, vLLM, llama.cpp, MLX, LM Studio. The upstream layer.
Did this app work for you on a specific rig? Submit the benchmark — it powers the model + hardware pages.