Live serving stats

GLM-5.3 Flash
5 queued
P50 queue wait: 3m 9s
DeepSeek V4.1 Flash
21 queued
P50 queue wait: 35s

P50 is the median time spent in the queue before processing, over the last 60 minutes.

API history

How to read these charts

Utilization uses the capacity available at each reading. Before API and Chat readings were separated on 29 October, the charts use combined usage. Gaps mean no reading.

GLM-5.3 Flash

OPERATIONAL HISTORY1-hour avg · Last sample 1m ago
API utilization
100%50%0%22 Nov, 17:3322 Nov, 20:3322 Nov, 23:33
Average API queue depth
116022 Nov, 17:3322 Nov, 20:3322 Nov, 23:33

DeepSeek V4.1 Flash

OPERATIONAL HISTORY1-hour avg · Last sample 1m ago
API utilization
100%50%0%22 Nov, 17:3322 Nov, 20:3322 Nov, 23:33
Average API queue depth
5025022 Nov, 17:3322 Nov, 20:3322 Nov, 23:33
All-time leadersTokens
01Navya Gupta16,886,412,526
02Udyaaa7,749,052,373
03deesh7,105,791,697
04akshaytheflash5,437,352,028
05Chakshita4,829,673,903
06Ayush Yadav4,012,655,794
07aadeshgupta29-cell3,992,267,878
08Pankaj Gupta3,368,271,545
Stewardship score
01Navya Gupta2,535
02Udyaaa2,080
03aadeshgupta29-cell1,876
04Sai Gautham Reddy Alavala1,853
05Pankaj Gupta1,716
06akshaytheflash1,625
07ArnavK-091,588
08deesh1,588

Tokens served

All time
93,617,395,565Input
713,623,185Output
94,331,018,750Total