Population: 1 model this gateway has served · metric: tokens · window: 30 days · normalization: none. These are this gateway's own requests, never a model's usage anywhere else. Models with no traffic are left out rather than ranked at zero, so the population here is smaller than the 1-model catalogue. The gateway has 2 days of daily buckets, so every window from here up contains the same rows and the totals do not move.
Name, context window, catalogue pricing, Intelligence index, Coding index and Agentic index are read from the API.
Requests served here, tokens served here and mean request latency are measured by this gateway from the requests it served, read back from POST /api/v1/analytics/query. They describe this deployment only — never a model’s usage anywhere else.
Week-over-week change is blank here — this gateway has not served the traffic it would count, and a count of nothing is left blank rather than shown as a zero.
Tokens this gateway served, in hour buckets. Total in window: 22.9M tokens.
Compare authors by the tokens this gateway served for them.
Throughput is blank here — the spec exposes it only to an authenticated caller, our capture was taken without a key, and nothing on this deployment has reported one yet. Nothing is generated to fill it.
Artificial Analysis coding index, read from each model's benchmarks block. Only models that advertise tools and carry a coding index appear — a model without one is left out rather than scored at zero.