GLM-5.3

zhipu/glm-5.3

The GLM-5.3 is part of the Zhipu GLM-5 series and is designed for complex reasoning, coding, and agent applications.

Real-Time Prices

Unit: USD per 1 million tokens (USD per 1M tokens). Prices are provided in real time by the gateway and settled based on actual token usage; usage that has already occurred is recorded at the price at the time it occurred and is not retroactively adjusted for price changes.

Input$0.04345 / 1M
Output$0.77 / 1M
Cache Hit$0.0429 / 1M
Reasoning$0.77 / 1M

Requests that fail and do not generate upstream output are not billed.View Plans and Pricing

Specifications and Capabilities

Context Window328K
Number of Abilities4 / 5
  • Stream OutputSSE returns results token by token, making it suitable for conversational interactionsSupport
  • Tool CallsFunction calls / tool usageSupport
  • Visual InputThis model does not support this capability.Not supported
  • Reasoning PatternsReasoning output that supports longer chains of thoughtSupport
  • JSON OutputStructured output, suitable for program parsingSupport

How to Call

Base URL: https://api.admesh.ai/v1, model name: zhipu/glm-5.3. Compatible with the OpenAI SDK and any client that supports custom Base URLs.

curl https://api.admesh.ai/v1/chat/completions \
  -H "Authorization: Bearer $ADMESH_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zhipu/glm-5.3",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

Note: The model name used in API calls is zhipu/glm-5.3 (with a slash), not the hyphenated form found in this page's URL. View the developer documentation

Other models from GLM

Models from the same company share the same access method; to switch between them, simply change the "model" field.

GLM-4.5-Air

zhipu/glm-4.5-air

The GLM-4.5-Air is the lightweight version of the GLM-4 series, suitable for cost-sensitive, high-frequency calls.

131KFlowToolsReasoning
Input$0.0715 / 1M
Output$0.4675 / 1M
Cache$0.01375 / 1M

GLM-4.6

zhipu/glm-4.6

GLM-4.6 is part of the Zhipu GLM-4 series and is suitable for Chinese conversation, writing, and tool invocation.

205KFlowToolsReasoning
Input$0.3025 / 1M
Output$1.21 / 1M
Cache$0.0605 / 1M

GLM-4.6V

zhipu/glm-4.6v

GLM-4.6V is GLM's visual understanding model, suitable for image-based question-answering and image-text hybrid tasks.

128KFlowToolsVisualReasoning
Input$0.165 / 1M
Output$0.495 / 1M
Cache$0.03025 / 1M

View All Plans and Rates

Start calling GLM-5.3

Once you register, you can create an API key. Usage is billed based on actual consumption, and the key is valid indefinitely. If you need a dedicated channel or a framework agreement, please contact us directly.