NVIDIALaunchAI ModelsOctober 14, 2026

NVIDIA Nemotron 3.5 Bolt 74B launches for RTX Spark local agents

Read original source →blogs.nvidia.com

The 74B-parameter, 9B-active MoE model ships Oct. 15 packed into 24GB via 2-bit quantization with quantization-aware training. It is purpose-built for agentic AI on RTX Spark laptops, which arrive Oct. 16 with up to 128GB unified memory and 1 petaflop of AI compute.

2 sources

More stories today

Open the live feed