Search results

GPU memory utilization needs to be set to higher value by default on IGX with 48 GB VRAM

Status: Open.

#28 In Project-MONAI/VLM-Surgical-Agent-Framework;

· mingxin-zheng opened on Feb 12, 2026

Using the Surgical Agent Framework on a new IGX box

Status: Open.

#26 In Project-MONAI/VLM-Surgical-Agent-Framework;

· mingxin-zheng opened on Feb 9, 2026

<code>--disable-mm-preprocessor-cache</code> is deprecated for vllm openai command line in docker settings

Status: Open.

#20 In Project-MONAI/VLM-Surgical-Agent-Framework;

· mingxin-zheng opened on Dec 26, 2025

<code>--enforce-eager</code> option slows down token generation

Status: Open.

#5 In Project-MONAI/VLM-Surgical-Agent-Framework;

· mingxin-zheng opened on May 9, 2025

<code>corrected_input</code> by SelectorAgent includes system messages and may be truncated

Status: Open.

#4 In Project-MONAI/VLM-Surgical-Agent-Framework;

· mingxin-zheng opened on Apr 27, 2025