Local worker interface

Compute Node

Configure availability, compute mode, and which model this node should keep warm.

Status

ConnectionConnected

AvailabilityAvailable

ComputeCUDA

Active jobs0

Coordinator node IDc03e83e5-5fa2-4c23-9c41-96a45d29a2bb

Compute mode Auto model type Model warming target

Auto picks the highest expected earning model for the selected type that fits this node, adjusted for how many nodes are already online per model.

Available for new work

NameNVIDIA GeForce GTX 1650

VendorNVIDIA

VRAM4096 MiB

Driver595.71.05

API URLhttp://central-brain:8000

Node namedocker-client-node

Auto typeembedding

Capacity1

Loaded modelBGE Large English v1.5