chiprook
← AI
AIOctober 14, 2026, 12:40

NVIDIA RTX Spark Launches Friday: 74B-Parameter AI Fits in 24GB Laptop

NVIDIA unveiled its RTX Spark platform built on a Grace CPU and Blackwell GPU superchip with up to 128GB of unified LPDDR5X memory and up to 1 petaFLOP of FP4 AI compute. The Nemotron 3.5 Bolt 74B model uses sparse mixture-of-experts (9B active parameters) and 2-bit quantization-aware training to fit in 24GB. The model releases October 15, with RTX Spark laptops on sale October 16.

NVIDIA RTX Spark Launches Friday: 74B-Parameter AI Fits in 24GB Laptop
#Nvidia#RTXSpark#Nemotron#MediaTek
Read next
AI

NVIDIA Releases Nemotron 3 Diarization: 100M-Parameter Open-Weight Model for 8 Speakers

AI

Perplexity’s local AI agent comes to Windows, but only for RTX GPUs with at least 24GB of VRAM

AI

NVIDIA's PivotOPD Teaches Multi-Turn AI Agents to Recover From Pivotal Mistakes

AI

Nvidia expands Southeast Asia AI deployments with Nemotron and Vera Rubin