Tarjei Mandt's picture

In a Training Loop 🔄

Tarjei Mandt

kernelpool

·

AI & ML interests

Everything MLX

Recent Activity

published a model 5 minutes ago

mlx-community/NVIDIA-Nemotron-3-Nano-4B-8bit

published a model 14 minutes ago

mlx-community/NVIDIA-Nemotron-3-Nano-4B-6bit

updated a model 16 minutes ago

mlx-community/NVIDIA-Nemotron-3-Nano-4B-4bit

View all activity

Organizations

upvoted an article 4 months ago

Article

Continuous batching from first principles

+1

Nov 25, 2025

•

344

upvoted a paper 5 months ago

Kimi Linear: An Expressive, Efficient Attention Architecture

Paper • 2510.26692 • Published Oct 30, 2025 • 132

upvoted a collection 6 months ago

DeepSeek-V3.2

4 items • Updated Dec 1, 2025 • 533