CoolFace
Modelpublic

adiagarwal/nanochat-models

sourceHugging Faceotherupdated 1mo agoView on Hugging Face
1likes
Model Card

NanoChat — On-Device Model Mirror

This repository hosts the quantized LiteRT model files used by [NanoChat](https://github.com/adiagarwalrock/NanoChat) — a free, open-source Android app that runs large language models fully on-device, with no internet connection required after the initial download and no API keys needed.

All models are mirrored here from their original sources so that NanoChat users on the Play Store can download them without needing a Hugging Face account.


📱 About NanoChat

NanoChat brings powerful, private AI chat to your Android phone. Everything runs locally on your device — your conversations never leave your phone.

  • —🔒 Fully private — no data ever sent to any server
  • —⚡ Fast — GPU-accelerated inference via Google's LiteRT runtime
  • —📦 No account or API key required — models download directly in-app
  • —🖼️ Multimodal — select models support image and audio input
  • —🆓 Free and open source

→ [View NanoChat on GitHub](https://github.com/adiagarwalrock/NanoChat)


🗂️ Repository Structure

Each model lives in its own folder:

nanochat-models/
├── Gemma3-1B-IT/
│   └── gemma3-1b-it-int4.litertlm
├── Gemma-3n-E2B-it/
│   └── gemma-3n-E2B-it-int4.litertlm
├── Gemma-3n-E4B-it/
│   └── gemma-3n-E4B-it-int4.litertlm
├── Phi-4-mini-instruct/
│   └── Phi-4-mini-instruct_multi-prefill-seq_q8_ekv4096.litertlm
├── Qwen2.5-1.5B-Instruct/
│   └── Qwen2.5-1.5B-Instruct_multi-prefill-seq_q8_ekv4096.litertlm
├── Qwen3.5-0.8B-Instruct/
│   └── model_multimodal.litertlm
├── Qwen3.5-2B-Instruct/
│   └── model_multimodal.litertlm
└── Qwen3.5-4B-Instruct/
    └── model_multimodal.litertlm

🤖 Available Models

💬 Chat Models

ModelSizeMin RAMVisionAudio
Gemma3-1B-IT558 MB6 GB❌❌
Qwen2.5-1.5B-Instruct1.49 GB6 GB❌❌
Qwen3.5-0.8B-Instruct1.16 GB4 GB✅❌
Qwen3.5-2B-Instruct2.78 GB8 GB✅❌
Gemma-3n-E2B-it3.40 GB8 GB✅✅
Phi-4-mini-instruct3.81 GB8 GB❌❌
Qwen3.5-4B-Instruct5.26 GB8 GB✅❌
Gemma-3n-E4B-it (experimental)4.58 GB12 GB✅✅

📥 Original Sources


⚖️ Licenses

Model FamilyLicense
Gemma 3, Gemma 3nGemma Terms of Use
Qwen 2.5, Qwen 3.5Apache 2.0
Phi-4-miniMIT
By downloading Gemma-family models through NanoChat you agree to the Gemma Terms of Use.

🔄 Keeping This Mirror Updated

This mirror is synced from the NanoChat `model_allowlist` whenever a new version ships. Only enabled models are mirrored. Disabled models are not uploaded but existing files are never deleted, ensuring backwards compatibility with older app versions.


Built with ❤️ for the NanoChat community.