Inference AtlasBETA
Checking status
API: checking
Model: unavailable
GPU: unavailable
INFRASTRUCTURE TELEMETRY

GPU dashboard

NVIDIA monitoring · Live readings when available · Updated every 2 seconds

GPU utilization— %

Unavailable values shown as —

VRAM used— GB

Unavailable values shown as —

Temperature— °C

Unavailable values shown as —

Active requests—

Unavailable values shown as —

VRAM total—
Queued requests—
Tokens generated—
Average TTFT—
Average latency—