A Self-hosted code completion assistant. from TabbyML. Teams that want private, self-hosted code completion instead of a cloud-only coding assistant.. Built for people who already know what they want and stopped tab-hopping.
An open-source, self-hosted AI coding assistant focused on code completion and private development environments.
Teams that want private, self-hosted code completion instead of a cloud-only coding assistant.
Users who want the broadest agentic coding workflow out of the box.
Best Match blends editorial suitability with feature fit, deployment fit, pricing, and supporting evidence.
User Love combines credible user and market signals with source diversity, sentiment, and evidence quality.
Momentum reflects recent launches, discussion, coverage, product updates, and source freshness.
Value compares pricing, free tier, local/open-source options, limits, and feature depth.
Evidence Level shows how much source support sits behind the recommendation.
Best Match blends editorial suitability with feature fit, deployment fit, pricing, and supporting evidence.
User Love combines credible user and market signals with source diversity, sentiment, and evidence quality.
Momentum reflects recent launches, discussion, coverage, product updates, and source freshness.
Value compares pricing, free tier, local/open-source options, limits, and feature depth.
Evidence Level shows how much source support sits behind the recommendation.
Tabby: Self-hosted AI coding assistant · 366 points · 150 comments
Show HN: Tabby – AI Coding Assistant Runs on Apple M1/M2 GPU · 3 points · 2 comments
33,564 stars · 1,748 forks · 317 open issues · 0 commits in 90d · active · license NOASSERTION · latest release v0.32.0
0 commits in the last 90 days · last push 2026-03-02 · latest release 2026-01-25 · repository active
Show HN: Tabby back end in 20 Python lines (self-hosted AI coding assistant) · 3 points · 2 comments
33,507 stars · 1,745 forks · 314 open issues · active · license NOASSERTION · latest release v0.32.0
Translation: Momentum means current activity, User Love is the reputation read, and Evidence Level tells you how much source support sits behind the recommendation.
No tracked updates yet. Subscribe and we will flag new product, pricing, and model changes as they land.