google-gemini / google-gemini/gemini-cli

Add support for local/offline language models (Ollola, LM Studio, etc.)

Open
#5,938 12 comments 46 reactions 0 assignees View on GitHub
area/agent kind/enhancement priority/p3 status/bot-triaged
Dominant language
TypeScript
Stars
107k
Forks
14.6k
Avg merge
2d 3h
Merged PRs (30d)
45

Description

### What would you like to be added?

Support for local language models like Ollama, LM Studio, and other offline AI models that can run on-premises without cloud connectivity.

### Why is this needed?

Enterprise organizations with strict data privacy requirements cannot use cloud-based AI services due to regulatory compliance (HIPAA, SOX, PCI-DSS, GDPR). Financial institutions, healthcare companies, government agencies, and defense contractors are prohibited from sending code or sensitive data to external cloud services.
Recent security incidents have made enterprises even more cautious about cloud AI:

The Samsung incident where employees accidentally leaked source code to ChatGPT
JPMorgan Chase restricting ChatGPT usage after code exposure concerns
Multiple healthcare organizations blocking AI tools due to HIPAA violations

Meanwhile, local models like CodeLlama, StarCoder, and DeepSeek-Coder have reached production quality and can run effectively on modern hardware. Tools like Ollama make local model deployment trivial, but there's no way to integrate them with Gemini CLI.
This creates a massive adoption barrier - entire enterprise development teams (thousands of potential users) cannot adopt Gemini CLI despite wanting to use it. Local model support would unlock this market while maintaining the superior UX that Gemini CLI provides.

### Additional context

_No response_

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.