llamaperf
Calculator
GPUs
Engines
Guides
Blog
Community
MCP
Advertise
Submit
Sign in
Submit
llamaperf
/
NVIDIA RTX 5090
/
report
GLM-5.2
on
NVIDIA RTX 5090
· llama.cpp
Tone: mixed
NVIDIA hardware
Aug 28, 2026
Share
X
Reddit
LinkedIn
Bluesky
Copy
link
System RAM
192 GB
VRAM reported
32 GB
Summary
User reports running GLM 5.2 on a system with 5x RTX Pro 6000 and 1x RTX 5090. User reports 98-99% task completion. User notes high heat and cost.
View source ↗
Feedback