CPU load9950X
—
32 threadsSampling
9X
9950X AI Portal
Local inference, services and hardware telemetry
Node online
System memoryRAM
—
——
GPU loadRadeon AI PRO R9700
—
——
Video memoryVRAM
—
——
Telemetry
Last 24 hours
Runtime control
AI deployments
One model at a time0
Performance
LLM benchmark
Select Bench on a running model
PP: 512, 2K, 8K, 32K prompt tokens · TG: 128, 256, 512 generated tokens on explanation, code, and story prompts. One request at a time; results use server timings where available.
No benchmark results yet.
Applications