NVIDIALaunchAI ModelsOctober 8, 2026

NVIDIA Nemotron 3.5 Bolt 74B-A9B ships for local agentic AI

Read original source →huggingface.co

A 74B-parameter, 9B-active MoE packed into 24GB via 2-bit quantization with quantization-aware training, released Oct. 15 for agentic AI on RTX Spark. NVIDIA's RTX Spark laptops, with up to 128GB unified memory and 1 petaflop of AI compute, launch Oct. 16.

2 sources

More stories today

Open the live feed