NVIDIA
Nemotron 3 Nano 4B
NVIDIA's efficient hybrid model for clear answers, coding, and instructions.
About Nemotron 3 Nano 4B
Nemotron 3 Nano is NVIDIA's efficiency-first line in the Nemotron 3 family, built for clear answers, coding and reasoning at low memory cost. This compact build targets phones that want NVIDIA-style quality without a flagship.
Every model in this catalog runs locally in the Offline AI Chat app. After the download finishes, Nemotron 3 Nano answers on your phone with no connection, and your prompts and replies are never uploaded.
- Download size
- 2.8 GB (Q4_K_M GGUF)
- Memory needed
- 6 GB minimum · 8 GB recommended
- Context in app
- 2,048 tokens
- Runs on
- Android 8+, 64-bit · iOS in private testing
- Access
- Offline Plus subscription
- Model source
- NVIDIA on Hugging Face