deeprelay requests get
Show one request by any id you hold for it
Show one request by any id you hold for it
Synopsis
Shows every recorded field of one request: ids, time, surface, model, the key (name, prefix and id), status and HTTP status, error code, finish reason, prompt, completion, cached and reasoning tokens, cost, whether your plan covered it, time to first token, total duration, and any request parameters the model ignored.
Pass whichever id you have — the server resolves them all to the same row:
- the
X-Request-Idresponse header (deeprelay_inf_…) - the response body id:
chatcmpl-…(chat completions),msg_…(messages),resp_…(responses) - a video id, or its stored request id (
deeprelay_med_…)
Backup models. When the request moved past the requested model to a backup (see deeprelay api-keys create), a Fallback line shows the requested model, the model that served (none when every model failed) and the reason, followed by the attempts trail: one line per model in the order tried, with its outcome (ok, failed or skipped), the error code or skip reason, how long it took, time to first token, and how many provider tries it used. A skipped model was never called, so it has no timing. The trail names models only, never providers. With -o json the same data is in requested_model, served_model, fallback_reason and attempts.
The id is sent exactly as you typed it. A request older than 90 days, or one made with a key you cannot see, reads as not found: the command prints request not found and exits 1. Fields that do not apply to the request are left out rather than shown as zero.
Examples
# By the id in a chat completion response body
deeprelay requests get chatcmpl-8f2c41d0a1b2
# By the X-Request-Id header, as JSON
deeprelay requests get deeprelay_inf_8f2c41d0a1b2 -o json
deeprelay requests get <request-id> [flags]
Options
-h, --help help for get
Options inherited from parent commands
--api-base string API base URL (override with DEEPRELAY_API_BASE env) (default "https://api.deeprelay.ai/v1")
--debug Enable debug logging to stderr
--no-preflight Skip the plan/credit check before inference requests (override with DEEPRELAY_NO_PREFLIGHT env)
-o, --output string Output format: table|json (default table on TTY, json otherwise) (default "table")
SEE ALSO
- deeprelay requests - Browse and export your request log