DeepSeek Provider

The DeepSeek provider offers high-performance chat and reasoning models with competitive pricing. node-llm supports both the DeepSeek-V3 chat model and the DeepSeek-R1 reasoning model.

Configuration

import { LLM } from "@node-llm/core";

LLM.configure({
  provider: "deepseek",
  apiKey: process.env.DEEPSEEK_API_KEY, // Optional if set in env
});

Specific Parameters

You can pass DeepSeek-specific parameters using .withParams().

const chat = LLM.chat("deepseek-chat")
  .withParams({ 
    presence_penalty: 0.5,
    frequency_penalty: 0.5,
    top_p: 0.9
  });

Features

• Models:
  • deepseek-chat: Optimized for speed and proficiency in broad tasks (DeepSeek-V3).
  • deepseek-reasoner: Optimized for complex reasoning and problem solving (DeepSeek-R1).
• Tools: Supported on deepseek-chat.
• Reasoning: Access inner thought process text from deepseek-reasoner.
• Streaming: Full streaming support for all models.
• Structured Output: Supported via automated prompt engineering and json_object mode transitions.

Usage Details

DeepSeek provides OpenAI-compatible endpoints, but node-llm handles the specific capability differences (like reasoning vs tool support) automatically through its internal registry.