Z.ai (Zhipu) · Flagship

GLM-5.3 API pricing and specs

Z.ai's flagship, the same base model as GLM-5.2 with better post-training, and thinking that cannot be turned off.

Input

⁨$1.40⁩

Cached input

⁨$0.26⁩

Output

⁨$4.40⁩

US dollars per million tokens, standard tier, short-prompt rate

Cached-input storage is free for a limited time. Requests with thinking disabled now fail, so remove that flag when migrating from GLM-5.2.

Spec sheet

API model string: glm-5.3
Provider: Z.ai (Zhipu)
Context: 1M tokens
Max output: —
Released: August 14, 2026
Batch API: No batch discount listed
Documented agent features: Tool callingAdjustable reasoning

What a task costs

Four reference workloads priced at this model's list rates, with caching applied after the first call. Your numbers will differ; the calculator lets you change every assumption.

Workload	Per task	Per 1,000 tasks
Support bot 4 calls, short tool lookups	⁨$0.035⁩	⁨$35⁩
Coding agent 30 calls, files and test output	⁨$1.63⁩	⁨$1,631⁩
Browser agent 25 calls, screenshots and page text	⁨$1.05⁩	⁨$1,052⁩
Research agent 15 calls, long search results	⁨$0.926⁩	⁨$926⁩