Get weekly model usage
Each public model's share of the tokens served in each ISO week (Monday 00:00 UTC start) for a stacked "top models" chart, plus this week's leaderboard. The chart draws the 10 models with the most tokens over the window and folds every other public model into other_share. The leaderboard lists the 20 models with the most tokens this week so far, with their rank and share, each compared with last week over the same span (Monday 00:00 UTC up to as_of), so a week in progress is not mistaken for a decline. change_pct is null until both spans have at least 20 requests. Same scope as /v1/rankings/models. Refreshed about once an hour.
weeks?integerNumber of weeks to chart, ending with the current week.
1 <= value <= 5212Successful response
application/json- response
as_of*stringWhen the numbers were computed; the current week and the leaderboard cover Monday 00:00 UTC up to this time.
date-timedata*array<>One entry per ISO week, oldest first, ending with the current week (in progress). Weeks without traffic are included with every share 0.
leaderboard*array<>Up to 20 models with the most tokens this week so far, sorted by rank.
weeks*integerNumber of weeks in data.
curl -X GET 'https://api.ohmygpt.com/v1/rankings/models/weekly'{ "as_of": "2019-08-24T14:15:22Z", "data": [ { "models": [ { "model": "string", "share": 0 } ], "other_share": 0, "week_start": "2019-08-24" } ], "leaderboard": [ { "change_pct": 0, "model": "string", "previous_rank": 1, "previous_share": 0, "rank": 1, "share": 0 } ], "weeks": 0}Get model historical stats GET
Returns historical latency, availability, tool-call, structured-output and effective user pricing statistics for a specific model, in two units: `auto` counts each request that left the provider choice to the gateway by the result the caller received, and each provider counts every upstream attempt it was sent. Only rates, percentiles, shares and prices are published, never request or token counts. Each number is null until it rests on enough samples: rates and shares need at least 20 samples, means, percentiles and weighted prices at least 30.
List models GET
Returns available models in an OpenAI-compatible format. Filters are limited to facts derivable from the catalog and enabled route pricing; category and arch are conservative derived mappings, so an unknown value is not treated as a match.