Get deployment metrics
curl --request GET \
--url https://api.veri.studio/v1/deployments/{deployment_id}/metrics \
--header 'Authorization: Bearer <token>'import requests
url = "https://api.veri.studio/v1/deployments/{deployment_id}/metrics"
headers = {"Authorization": "Bearer <token>"}
response = requests.get(url, headers=headers)
print(response.text)const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};
fetch('https://api.veri.studio/v1/deployments/{deployment_id}/metrics', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"total_requests": 123,
"total_prompt_tokens": 123,
"total_completion_tokens": 123,
"total_cost_usd": 123,
"avg_latency_ms": 123,
"p50_latency_ms": 123,
"p90_latency_ms": 123,
"p95_latency_ms": 123,
"p99_latency_ms": 123,
"error_rate": 123,
"uptime_seconds": 123
}Authorizations
API key with the vk_ prefix. Create one from the dashboard.
Path Parameters
Deployment ID
Response
Uptime, cost, lifetime request counters, and past-24-hour latency and error stats
Lifetime count of your API requests (eval calls excluded).
Lifetime prompt tokens of your API requests (eval calls excluded).
Lifetime completion tokens of your API requests (eval calls excluded).
GPU cost accrued over the deployment's life: every replica of every
boot, from its launch to its termination (or now), at its per-box rate.
This is what billing charges, before rounding to the cent; the
deployment's own total_cost_usd is the part already settled.
Mean end-to-end latency of your API requests in the past 24 hours;
null when there were none. The percentiles and error_rate cover the
same window.
Seconds during which at least one GPU box was up and billed, summed over every boot (wakes included, parked time excluded). Null until a box launches.
curl --request GET \
--url https://api.veri.studio/v1/deployments/{deployment_id}/metrics \
--header 'Authorization: Bearer <token>'import requests
url = "https://api.veri.studio/v1/deployments/{deployment_id}/metrics"
headers = {"Authorization": "Bearer <token>"}
response = requests.get(url, headers=headers)
print(response.text)const options = {method: 'GET', headers: {Authorization: 'Bearer <token>'}};
fetch('https://api.veri.studio/v1/deployments/{deployment_id}/metrics', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));{
"total_requests": 123,
"total_prompt_tokens": 123,
"total_completion_tokens": 123,
"total_cost_usd": 123,
"avg_latency_ms": 123,
"p50_latency_ms": 123,
"p90_latency_ms": 123,
"p95_latency_ms": 123,
"p99_latency_ms": 123,
"error_rate": 123,
"uptime_seconds": 123
}