BigHugger
GH Repository · off-grid-ai

OGAM

The Swiss Army Knife of Offline AI. Chat, see, speak, and generate images on your phone or Mac — GGUF LLMs, vision, Whisper speech-to-text, Stable Diffusion, tool calling, and local-network servers. Runs on your CPU, GPU, or NPU. No account, no API key, zero data leaves your device.

stars
3,117
30-day movement
+196/day
Related entries
60
Connections
4
typescriptrubyTypeScriptvision-language-modelnodemcpmacosiosandroidprivacy-firstmobile-aiwhisper-cppggufllama-cpplocal-aiedge-aioffline-llmreact-nativewhisper-androidstable-diffusion-androidoffline-aitool-callingondevicemcp-server

OGAM is a cross-platform (Android, iOS, macOS) offline AI app and MCP server that runs GGUF LLMs, vision models, Whisper speech-to-text, and Stable Diffusion locally on CPU, GPU, or NPU. It supports chat, tool calling, image generation, and exposing local-network servers, with no accounts or API keys required.

You want capable AI chat, vision, speech, and image generation running entirely on your own device with zero data leaving it.

Use it to

  • Chat with local GGUF LLMs on a phone
  • Run Whisper speech-to-text offline
  • Generate images with on-device Stable Diffusion
  • Serve AI tools to local-network clients via MCP
  • Integrate local tool calling into agent workflows

For Developers and privacy-focused users of AI on mobile or Mac

Role
mcp-server
Language
TypeScript
Licence
MIT
Forks
298
Open issues
136
Last push
2026-09-16
Latest release
v0.0.2 · 2026-02-07
topicsoffline-aiggufwhisperstable-diffusionmcpondevice