@ctnzr
Announcing NVIDIA Nemotron 3 Super! π120B-12A Hybrid SSM Latent MoE, designed for Blackwell π36 on AAIndex v4 πup to 2.2X faster than GPT-OSS-120B in FP4 πOpen data, open recipe, open weights Models, Tech report, etc. here: https://t.co/CAYpP1iK3i And yes, Ultra is coming! https://t.co/QuguMQaC8S