TabbyML is an open-source, self-hosted coding assistant built by two ex-Googlers that generates code and can be customized for enterprise use. Launched in April, the project has been starred roughly 11,000 times on GitHub. The company emphasizes self-hosting to allow enterprises to incorporate proprietary code and to fine-tune models from user edits. TabbyML aims to lower deployment barriers by recommending models trained on one-to-three billion parameters, trading short-term quality for lower serving costs. Co-founders Meng Zhang and Lucy Gao argue that as compute costs fall and open-source model quality improves, incumbents like GitHub Copilot will lose advantage. TabbyML has secured $3.2 million in seed funding to develop its open-source code generator, with investors Yunqi Partners and ZooCap participating.

Total raised

$3M

Funding rounds

1

Latest round

Seed

Latest activity

Oct 2023