Model vs model

glm-5.3 vs kimi-k3

Both models on Cent AI, priced from the active price version, next to each publisher's official rate.

glm-5.3

Checking status90% off

Price version 3 · in effect since Oct 30, 2026

per 1M tokens

InputOfficial price $1.40$0.14
Cache readOfficial price $0.26$0.014
Cache write$0.14
OutputOfficial price $4.40$0.44

Minimum per request
$0.00002
Context window
—

Endpoints

  • /v1/chat/completions
  • /v1/responses
  • /v1/messages
glm-5.3 model page →

kimi-k3

Checking status79% off

Price version 3 · in effect since Oct 31, 2026

per 1M tokens

InputOfficial price $3.00$0.63
Cache readOfficial price $0.30$0.063
Cache writeOfficial price $3.00$0.63
Cache write 1 hOfficial price $6.00$0.63
OutputOfficial price $15.00$3.15

Minimum per request
$0.00002
Context window
—

Endpoints

  • /v1/chat/completions
  • /v1/responses
  • /v1/messages
kimi-k3 model page →

The same workload on both

Z.aiglm-5.3Official price $143.00$14.30
Moonshot AIkimi-k3Official price $312.00$65.52

On this workload glm-5.3 costs $51.22 less than kimi-k3 at Cent AI.

Workload: 99,000,000 input and 1,000,000 output tokens a month, no cache. With cache the gap changes; try your own mix in the calculator.

Try glm-5.3 and kimi-k3 with one key

One prepaid balance for every model. Pay with QRIS or USDT/USDC.

Start free