The Zhipu adapter provides integration with the Zhipu AI API. Zhipu AI offers the GLM series of large language models, with an API that is fully compatible with the OpenAI format.

Installation

pnpm add @amux.ai/llm-bridge @amux.ai/adapter-zhipu

Basic Usage

import { createBridge } from '@amux.ai/llm-bridge'
import { zhipuAdapter } from '@amux.ai/adapter-zhipu'

const bridge = createBridge({
  inbound: zhipuAdapter,
  outbound: zhipuAdapter,
  config: {
    apiKey: process.env.ZHIPU_API_KEY
  }
})

const response = await bridge.chat({
  model: 'glm-4.7',
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'What is Amux?' }
  ]
})

console.log(response.choices[0].message.content)

Supported Models

Zhipu AI offers various GLM models:

Model	Description
glm-4.7	Latest flagship model with best performance
glm-4.6	High-performance model balancing capability and cost
glm-4.5	Cost-effective model for everyday tasks
glm-4v	Multimodal model supporting visual understanding

Zhipu AI models are continuously updated. Please refer to the official documentation for the latest model list.

Key Features

Basic Chat

const response = await bridge.chat({
  model: 'glm-4.7',
  messages: [
    { role: 'user', content: 'Tell me about Zhipu AI' }
  ],
  temperature: 0.7,
  max_tokens: 1000
})

Function Calling

const response = await bridge.chat({
  model: 'glm-4.7',
  messages: [
    { role: 'user', content: 'What time is it in Beijing?' }
  ],
  tools: [{
    type: 'function',
    function: {
      name: 'get_current_time',
      description: 'Get the current time for a specified city',
      parameters: {
        type: 'object',
        properties: {
          city: { type: 'string', description: 'City name' }
        },
        required: ['city']
      }
    }
  }]
})

Streaming

const stream = bridge.chatStream({
  model: 'glm-4.7',
  messages: [
    { role: 'user', content: 'Tell me a story' }
  ],
  stream: true
})

for await (const chunk of stream) {
  if (chunk.choices[0]?.delta?.content) {
    process.stdout.write(chunk.choices[0].delta.content)
  }
}

Configuration Options

const bridge = createBridge({
  inbound: zhipuAdapter,
  outbound: zhipuAdapter,
  config: {
    apiKey: process.env.ZHIPU_API_KEY,
    baseURL: 'https://open.bigmodel.cn/api/paas', // Default value
    timeout: 60000
  }
})

Feature Support

Feature	Supported	Notes
Chat Completion	✅	Fully supported
Streaming	✅	Fully supported
Function Calling	✅	Fully supported
Vision	✅	glm-4v supported
System Prompt	✅	Fully supported
JSON Mode	✅	Fully supported
Web Search	✅	Supported

Best Practices

1. Choose the Right Model for Your Needs

// Use cost-effective model for everyday tasks
const quickResponse = await bridge.chat({
  model: 'glm-4.5',
  messages: [{ role: 'user', content: 'Hello' }]
})

// Use flagship model for complex tasks
const complexTask = await bridge.chat({
  model: 'glm-4.7',
  messages: [
    { role: 'user', content: 'Please analyze the performance issues in this code...' }
  ]
})

2. Use System Prompts to Optimize Output

const response = await bridge.chat({
  model: 'glm-4.7',
  messages: [
    {
      role: 'system',
      content: 'You are a professional technical documentation writer. Please answer questions in concise, professional language.'
    },
    {
      role: 'user',
      content: 'What is a RESTful API?'
    }
  ]
})

3. Handle Multi-turn Conversations

const messages = [
  { role: 'user', content: 'First question' },
  { role: 'assistant', content: 'First answer' },
  { role: 'user', content: 'Follow-up question' }
]

const response = await bridge.chat({
  model: 'glm-4.7',
  messages
})

Converting with OpenAI

Zhipu AI is fully compatible with the OpenAI format:

import { openaiAdapter } from '@amux.ai/adapter-openai'
import { zhipuAdapter } from '@amux.ai/adapter-zhipu'

const bridge = createBridge({
  inbound: openaiAdapter,
  outbound: zhipuAdapter,
  config: {
    apiKey: process.env.ZHIPU_API_KEY
  }
})

// Send requests using OpenAI format
const response = await bridge.chat({
  model: 'gpt-4',
  messages: [{ role: 'user', content: 'Hello' }]
})

Zhipu Adapter