Local deployment of Tabby requires the followingstandard process(based on the latest v0.24.0 release):
- environmental preparation::
- Installing Docker version 20.10+
- NVIDIA drivers and CUDA Toolkit recommended (version 11.8/12.x recommended to enable GPU acceleration)
- Ensure disk space ≥ 10GB
- Mirror Image Acquisition: Implementation
docker pull tabbyml/tabbyPull the latest mirrors - service activation: Copy the following commands and adjust the parameters according to the hardware:
docker run -it --gpus all -p 8080:8080 -v $HOME/.tabby:/data tabbyml/tabby serve --model TabbyML/StarCoder-1B --device cuda --chat-model Qwen2-1.5B-Instruct- removes
--gpus allCPU-only operation possible - First startup takes 5-10 minutes to download the model
- removes
- Verification runs: Access
http://localhost:8080View Welcome Page
Key caveat: data is persistently stored in the~/.tabbyCatalog; available through--parallelism 4Enhance concurrency performance; if deployed in enterprise servers need to configure the reverse proxy.
This answer comes from the articleTabby: a native self-hosted AI programming assistant that integrates into VSCodeThe
May not be reproduced without permission:AI productivity tools " How to deploy Tabby services locally via Docker?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil