SelfHostMap

Self-hosted coding assistant for your dev team (Tabby)

Quick answer

Tabby serves code completion and chat to IDE extensions from your own server and can index your repositories for context. It can run its own models or call models served by Ollama over HTTP for completion, chat and embeddings. Tabby needs a GPU server for practical use; we do not estimate GPU prices because they vary too much by card and provider.

Last updated · sizing from each app's official docs · prices from Hetzner Cloud (2026-10-22)

Components

  • Tabby: Completion and chat server for IDEs, repository indexing
    Official hardware: Not published · License: Apache-2.0 (ee directory under a separate license) (Open core)
  • Ollamaoptional: Serve completion/chat/embedding models over HTTP
    Official hardware: Not published (measured by us: 1.9 GiB RAM while working) · License: MIT (Permissive)

Combined hardware and monthly cost

Required components, official minimumsNeeds a GPU server (Tabby); not estimated.
Smallest server that fitsa GPU server, which we do not price
Measured RAM in our own testOllama 0.40.8: 1.9 GB peak while working. Measured on SelfHostMap test VM (Oct 2026), 2026-10-22; container RAM only, not the whole server; no load test. Model-runner figures are for the small test models named on the app page; larger models need far more RAM.
Including optional componentsNeeds a GPU server (Tabby); not estimated.

How we got this: we added up the official minimum vCPU and RAM each project publishes (linked on its app page) and took the cheapest Hetzner Cloud Regular Performance (shared AMD vCPU) plan that meets the total. Prices read from www.hetzner.com/cloud/regular-performance on 2026-10-22. "From" monthly price cap shown by Hetzner, excluding VAT. Prices vary by location. Covers the server only; it excludes LLM API usage, backups, extra storage, GPUs and your time. Running everything on one server is our simplifying assumption. Measured figures, where we have them, are on each app page.

Cost vs GitHub Copilot

1. The SaaS you pay for

2. Self-hosted stack

3. Model API spend (self-hosted side)
SaaS per year—
Self-host per year—
Break-even—

SaaS (cumulative)Self-host incl. setup (cumulative)
Open in full calculator
Assumptions and sources

    Deployment outline

    1. Install the NVIDIA Container Toolkit on a GPU host, then run the official Tabby CUDA image. source
    2. Optional: point Tabby at Ollama in ~/.tabby/config.toml ([model.completion.http] kind = "ollama/completion", and openai/chat for chat). source
    3. Check VRAM against the models you choose (CodeLlama-7B in int8 needs about 8 GB). source

    This is a summary of the official docs, not a full tutorial. Each app page has the exact official install command.

    Gotchas

    • Tabby: The CUDA image needs the NVIDIA Container Toolkit installed on the host. source
    • Tabby: On SELinux systems you may need to add the :Z option to the volume mount. source
    • Ollama: Older NVIDIA cards (compute capability 5.0–6.2) need driver 570 or newer. source
    • Ollama: With the Vulkan backend, Ollama needs root or the cap_perfmon capability (sudo setcap cap_perfmon+ep /usr/local/bin/ollama) to read free VRAM; otherwise it schedules models using approximate sizes. source

    What this does not replace

    • Tabby needs a GPU server for practical use; we do not estimate GPU server cost.
    • Copilot's cloud agent, pull-request code review and choice of hosted frontier models are not matched.

    FAQ

    Do I need a GPU for a self-hosted coding assistant?

    For Tabby, practically yes. The project recommends an NVIDIA GPU, and VRAM depends on the models you pick. We do not estimate GPU server prices.

    How much does a local coding assistant (github copilot alternative) stack cost per month?

    Tabby needs a GPU server for practical use; we do not estimate GPU prices because they vary too much by card and provider.

    What hardware does Tabby need?

    Needs a GPU server (Tabby); not estimated..

    Can a business use these apps commercially?

    Tabby: Apache-2.0 (ee directory under a separate license) (internal business use allowed). Ollama: MIT (internal business use allowed). Summaries only, not legal advice.

    Is it cheaper than GitHub Copilot?

    It depends on team size, API usage and how you value your time. Use the calculator on this page; it starts with 10 users and GitHub Copilot Business at the price published on 2026-10-22.

    Other stacks

    Not sure which app fits? Try the 5-question wizard.

    Every figure comes from the linked app pages and official docs. Spotted something wrong? Tell us.