High-resolution Vision-Language Models (VLMs) offer impressive accuracy but come with significant computational costs—processing thousands of tokens per image can consume 5GB of GPU memory and add 15 seconds of latency. The HiRED (High-Resolution Early Dropping) framework addresses this challenge by intelligently selecting only the most informative visual tokens based on attention patterns. By keeping just 20% of tokens, researchers achieved a 4.7× throughput increase and 78% latency reduction while maintaining accuracy across vision tasks. This research, conducted on Chameleon's infrastructure using RTX 6000 and A100 GPUs, demonstrates how thoughtful optimization can make advanced AI more accessible and affordable.

Learn how to leverage Trovi's new GitHub integration to easily create and update reproducible research artifacts. This step-by-step guide shows you how to configure your GitHub repository with RO-crate metadata and import it directly into Trovi, enabling better collaboration and adherence to FAIR principles for your experiments.

This month, we have reminders for KVM@TACC and CHI@Edge outages later this month. Additionally, we have version 1.1 of python-chi, and improvements to reservations!