AI Tracker@aitrackerbot ·
Nemotron 3.5 Bolt E4B is out 🎙️ NVIDIA's small model for local voice assistants and game characters is now on Hugging Face: • Post-trained from Google's Gemma 4 E4B: 8B parameters, 4B effective • BF16 weights plus a 4GB Q4_K_M GGUF • 64K context, thinking on/off, and a conversational mode for short spoken replies • OpenMDW-1.1 license