Report Training Metrics and Status - Node Agent

Node agents push training telemetry and status transitions to the PIN control plane. Two endpoints cover this: /v1/agent/metrics for per-step numeric metrics and /v1/agent/status for job status transitions.

Both endpoints require X-Agent-Secret. They are meant for the node agent, not for end users.

Push metrics

POST /v1/agent/metrics

Request body

string

required

Sending agent identifier.

string

required

Job the metrics belong to.

integer

required

Training step index.

number

Training loss.

number

Reward (RL jobs).

number

Tokens per second.

number

GPU utilization percent.

Push status

POST /v1/agent/status

Request body

string

required

Sending agent identifier.

string

required

Job identifier.

string

required

New status: running, paused, completed, failed.

string

Error message when status = failed.

Response

Both endpoints return {"code": 0, "data": {"ok": true}, "message": ""} on success.

Example

curl -X POST https://ariapin.example.com/v1/agent/metrics \
  -H "X-Agent-Secret: $AGENT_SECRET" \
  -H "Content-Type: application/json" \
  -d '{"agent_id":"ag_01","job_id":"job_0198a","step":250,"loss":0.42,"throughput":1850,"gpu_util":92}'

PIN API

ROUTER API

AGENT API

Report Training Metrics and Status - Node Agent

Push metrics

Request body

Push status

Request body

Response

Example

​Push metrics

​Request body

​Push status

​Request body

​Response

​Example

Push metrics

Request body

Push status

Request body

Response

Example