zhipu-text

Zhipu AI Text Generation Models (Text Generation)

Model Overview

Model	Model ID	Parameters	Context
GLM-5	`glm-5`	744B / 40B active	128K
GLM-4.7	`glm-4.7`	-	128K
GLM-4.6	`glm-4.6`	355B / 32B active	128K
GLM-4.5 Series	`glm-4.5` and 5 variants	-	128K
GLM-4.7-FlashX	`glm-4.7-flashx`	-	128K
GLM-4.7-Flash	`glm-4.7-flash`	30B	200K

Selection Recommendations

Flagship: glm-5 — Agentic Engineering positioning
Cost-Effective Flagship: glm-4.7 — No. 1 open source on Code Arena
High-Speed: glm-4.7-FlashX
Stable: glm-4.6
Flexible: glm-4.5 Series — 5 variants covering different performance and cost requirements
Free: glm-4.7-flash — Free and has a 200K context

Shared Capabilities

All text models support: thinking mode (thinking.type: enabled/disabled), streaming, Function Calling, context caching, structured JSON Schema output, and MCP support.

API Key Configuration

Load from ~/.glm/.env , use ZHIPU_API_KEY. Key acquisition address:https://open.bigmodel.cn/usercenter/apikeys

API Invocation

POST https://open.bigmodel.cn/api/paas/v4/chat/completions

Auth: Bearer token，Content-Type: application/json

The request body includes:model, messages, stream, temperature, max_tokens

Thinking mode: "thinking": {"type": "enabled"}

SDK Example

from zai import ZhipuAI
client = ZhipuAI(api_key="your-key")
response = client.chat.completions.create(
    model="glm-5",
    messages=[{"role": "user", "content": "你好"}]
)

Recommended Scenarios

Agentic Coding, long-chain autonomous agents, office automation, role-playing/script generation, professional translation, structured data extraction, customer service quality inspection.

About this skill