The DeepSeek adapter provides integration with the DeepSeek API. DeepSeek API is fully compatible with OpenAI format and offers high-value AI services.

Installation

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

Basic Usage

import { createBridge } from '@amux.ai/llm-bridge'
import { deepseekAdapter } from '@amux.ai/adapter-deepseek'

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

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

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

Supported Models

Model	Description	Features
`deepseek-chat`	General chat model	High value, suitable for daily conversations
`deepseek-coder`	Code-specialized model	Focused on code generation and understanding
`deepseek-reasoner`	Reasoning model	Advanced reasoning capabilities

Key Features

Code Generation

DeepSeek Coder is optimized for coding tasks:

const response = await bridge.chat({
  model: 'deepseek-coder',
  messages: [
    {
      role: 'user',
      content: 'Write a quicksort function in TypeScript'
    }
  ],
  temperature: 0.3 // Lower temperature for more deterministic code
})

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

Function Calling

const response = await bridge.chat({
  model: 'deepseek-chat',
  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' }
        },
        required: ['city']
      }
    }
  }]
})

Streaming

const stream = bridge.chatStream({
  model: 'deepseek-chat',
  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: deepseekAdapter,
  outbound: deepseekAdapter,
  config: {
    apiKey: process.env.DEEPSEEK_API_KEY,
    baseURL: 'https://api.deepseek.com', // Default
    timeout: 60000
  }
})

Converting with OpenAI

DeepSeek is fully compatible with OpenAI format:

import { openaiAdapter } from '@amux.ai/adapter-openai'
import { deepseekAdapter } from '@amux.ai/adapter-deepseek'

// OpenAI format → DeepSeek API
const bridge = createBridge({
  inbound: openaiAdapter,
  outbound: deepseekAdapter,
  config: {
    apiKey: process.env.DEEPSEEK_API_KEY
  }
})

// Send request in OpenAI format
const response = await bridge.chat({
  model: 'gpt-4', // Will be mapped to deepseek-chat
  messages: [{ role: 'user', content: 'Hello' }]
})

Feature Support

Feature	Supported	Notes
Chat Completion	✅	Fully supported
Streaming	✅	Fully supported
Function Calling	✅	Fully supported
Vision	❌	Not supported
System Prompt	✅	Fully supported
Reasoning	✅	deepseek-reasoner model
JSON Mode	✅	Structured output

Advantages

High Value: Much cheaper than GPT-4 with comparable performance
Strong Coding: DeepSeek Coder excels at coding tasks
OpenAI Compatible: Can seamlessly replace OpenAI API
Chinese Friendly: Excellent Chinese language support

Best Practices

1. Choose the Right Model

// Daily conversations use deepseek-chat
const chatResponse = await bridge.chat({
  model: 'deepseek-chat',
  messages: [{ role: 'user', content: 'Tell me about yourself' }]
})

// Coding tasks use deepseek-coder
const codeResponse = await bridge.chat({
  model: 'deepseek-coder',
  messages: [{ role: 'user', content: 'Write a binary search algorithm' }]
})

// Complex reasoning use deepseek-reasoner
const reasoningResponse = await bridge.chat({
  model: 'deepseek-reasoner',
  messages: [{ role: 'user', content: 'Solve this complex problem...' }]
})

2. Optimize Code Generation

const response = await bridge.chat({
  model: 'deepseek-coder',
  messages: [
    {
      role: 'system',
      content: 'You are a professional programmer. Generate clean, efficient, well-commented code.'
    },
    {
      role: 'user',
      content: 'Implement an LRU cache in Python'
    }
  ],
  temperature: 0.2, // Low temperature for more deterministic code
  max_tokens: 1000
})

DeepSeek Adapter