Get Gpu Probe Status

GET

/api/settings/gpu-probe-status

• DivinityEngine - Local development
• DivinityEngine - Localhost alternative

Return the cached DirectML GPU probe result.

The probe runs during backend startup to test whether DirectML inference works on this GPU without segfaulting. The result is cached to disk.

Responses

200

Successful Response

GpuProbeStatusResponse

Response for the GPU probe status endpoint.

object

status

required

Status

Probe result: pass, fail, crash, unavailable, not_run

string

gpuName

Any of:

string

string

null

null

ortVersion

Any of:

string

string

null

null

message

required

Message

Human-readable status message

string