All systems operational

pg.ai gateway · read 2026-09-24T07:03:51.845Z · 2026-09-24T03:03:51-04:00

Upstream providers

Meta MuseOperationalAnswering. The scheduled check succeeded.https://meta.pg.ai.ajam.dev · a scheduled check less than a minute ago2026-09-24T07:03:30.686Z · 2026-09-24T03:03:30-04:00
ZAI GLMOperationalAnswering. The scheduled check succeeded.https://zai.pg.ai.ajam.dev · a scheduled check less than a minute ago2026-09-24T07:03:48.331Z · 2026-09-24T03:03:48-04:00

This gateway

APIOperationalServing. This page is answered by the same Worker as the API.
Key storeOperationalReadable. Issued keys can be verified.
Model cacheOperationalReadable. Model listings are served from cache.
D1OperationalAnswered in 244 ms.
KVOperationalAnswered in 104 ms.

Remaining quota

Meta Muse · 5 hours100% leftResets in 34 minutes. Read less than a minute ago.2026-09-24T07:38:28.000Z · 2026-09-24T03:38:28-04:00
Meta Muse · Weekly2% leftResets in 3 d 16 h. Read less than a minute ago.2026-09-28T00:00:00.000Z · 2026-09-27T20:00:00-04:00
ZAI GLM · 5 hours99.5% leftResets in 15 minutes. Read less than a minute ago.2026-09-24T07:19:32.239Z · 2026-09-24T03:19:32-04:00
ZAI GLM · Weekly39.2% leftResets in 1 d 5 h. Read less than a minute ago.2026-09-25T12:10:01.949Z · 2026-09-25T08:10:01-04:00

Each figure is what is left at the highest usage seen since that window reset, which is the conservative half.

These are the whole gateway's shared allowances. They are not your key's: a pool with room does not mean a key has any. Read GET /v1/usage with your own key for that.

Usage

Today so far: 719 requests, 19,877,000 tokens. Last 7 days: 39,692 requests, 2,947,957,018 tokens, over the 5 days this record covers; 4 of those days recorded any tokens.

Recorded since 2026-09-20: the 2 earlier days in this window are not recorded, which is not the same as a day with no traffic. Every day and hour is UTC.

2026-09-18—not recorded
2026-09-19—not recorded
2026-09-20182requests0tokens
2026-09-218,987requests699,272,215tokens
2026-09-2214,899requests1,629,459,143tokens
2026-09-2314,905requests599,348,660tokens
2026-09-24719requests19,877,000tokens

Requests by hour of day

Busiest hour: 17:00 UTC · 13:00 America/New_York, 8,047 requests over the window.

00· 20:00 America/New_York — 569 requests
01· 21:00 America/New_York — 1,226 requests
02· 22:00 America/New_York — 967 requests
03· 23:00 America/New_York — 1,185 requests
04· 00:00 America/New_York — 4,477 requests
05· 01:00 America/New_York — 1,570 requests
06· 02:00 America/New_York — 2,949 requests
07· 03:00 America/New_York — 777 requests
08· 04:00 America/New_York — 762 requests
09· 05:00 America/New_York — 3,882 requests
10· 06:00 America/New_York — 958 requests
11· 07:00 America/New_York — 2,301 requests
12· 08:00 America/New_York — 662 requests
13· 09:00 America/New_York — 637 requests
14· 10:00 America/New_York — 795 requests
15· 11:00 America/New_York — 1,138 requests
16· 12:00 America/New_York — 3,988 requests
17· 13:00 America/New_York — 8,047 requests
18· 14:00 America/New_York — 933 requests
19· 15:00 America/New_York — 656 requests
20· 16:00 America/New_York — 386 requests
21· 17:00 America/New_York — 303 requests
22· 18:00 America/New_York — 257 requests
23· 19:00 America/New_York — 267 requests

These are totals for the whole gateway. They are not a quota and they do not say what any key may still spend: read GET /v1/usage with your own key for that.

Published tariff windows

ZAI GLMPeak nowZ.AI's GLM Coding Plan charges off-peak usage at half the credit rate; peak is Monday to Friday, 06:00 to 10:00 UTC. Everything else is off-peak.https://docs.z.ai/devpack/notice/usage-revision (read 2026-09-19)

These are the providers' own published windows, not this deployment's.