metrics

Range-query metrics (PromQL)

PromQL over your org's metrics: platform telemetry, free and automatic - GPU (gmn_gpu_utilization, gmn_gpu_memory_used_bytes, gmn_gpu_power_watts, gmn_gpu_temperature_celsius), system per container (gmn_cpu_seconds_total, gmn_memory_used_bytes, gmn_memory_limit_bytes, gmn_network_receive_bytes_total, gmn_network_transmit_bytes_total), and spend (gmn_node_spend_usd, gmn_org_mtd_spend_usd) - labeled {node, mission} on nodes and {run, mission} on jobs, plus everything your registered scrapes collect. Range query; results are bounded and downsampled (at most 20 series x 200 points, truncated:true when cut) - ask narrower questions rather than paging. $mission/$run in the query substitute the matching params.

post/preview/metrics/queries

Request

  • Base URL: https://autoresearch.sfcompute.com
  • URL: https://autoresearch.sfcompute.com/preview/metrics/queries
  • Auth: HTTP bearer

Query parameters

workspacestring

Act in this workspace (its slug). Omitted: the token's bound workspace, then the organization's default. Never ambient - there is no switch-workspace mutation on this door.

Request body

endstring nullable

Range end, RFC3339 (default: now).

missionstring nullable

Substituted for $mission in the query.

querystring required

PromQL. $mission and $run substitute the params below, e.g. sum(rate(vllm:generation_tokens_total{mission="$mission"}[5m])).

runstring nullable

Substituted for $run in the query (a job-/cmd-/ses- scope id).

startstring nullable

Range start, RFC3339 (default: one hour before end).

step_secondsinteger nullable

Step seconds (default: the range divided into ~200 points).

Response

The capability's JSON result.

objectstring

The resource type, when the result is one resource.

Changes

No recorded changes to this operation across all 1 revision of this API.