TABBY

Open-source, self-hosted AI coding assistant

Apache-2.0

ABOUT

Teams that want AI coding assistance often have to send proprietary code to third-party hosted services or accept limited control over models, infrastructure, and data handling. Tabby provides a self-hosted coding assistant with code completion and related coding workflows so organizations can run an internal AI developer tool on their own infrastructure.

INSTALL

docker run -d \
  --name tabby \
  --gpus all \
  -p 8080:8080 \
  -v $HOME/.tabby:/data \
  registry.tabbyml.com/tabbyml/tabby \
    serve \
    --model StarCoder-1B \
    --chat-model Qwen2-1.5B-Instruct \
    --device cuda

INTEGRATION GUIDE

1. Run self-hosted code completion for developers 2. Deploy an internal AI coding assistant for a team 3. Use inline chat and answer engine inside coding workflows 4. Connect IDE extensions to a private coding backend 5. Run supported coding and chat models on local infrastructure

TABBY

ABOUT

INTEGRATION GUIDE

TAGS