Skip to content

DeepSeek

Overview

DeepSeek builds high-performance language models with strong reasoning capabilities, including the chain-of-thought DeepSeek Reasoner. Floopy proxies requests to DeepSeek’s OpenAI-compatible API and surfaces reasoning token metrics via response headers.

Supported Models

ModelContext WindowNotes
deepseek-chat64KGeneral-purpose chat model
deepseek-reasoner64KChain-of-thought reasoning model

Setup

  1. Go to Settings > Providers in the dashboard.
  2. Click Add provider and select DeepSeek.
  3. Paste your DeepSeek API key and click Save.

Usage

import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.floopy.ai/v1",
apiKey: process.env.FLOOPY_API_KEY,
});
const response = await client.chat.completions.create({
model: "deepseek-chat",
messages: [{ role: "user", content: "Explain quantum computing." }],
});

Provider-Specific Features

  • Reasoning tokens — When using deepseek-reasoner, Floopy returns a Floopy-Reasoning-Tokens response header with the number of tokens used for chain-of-thought reasoning, separate from the completion token count.
  • Chain-of-thought — DeepSeek Reasoner performs multi-step reasoning before producing its final answer, making it well suited for math, logic, and complex analysis tasks.

Fallback

Route to OpenAI if DeepSeek is unavailable:

Terminal window
curl https://api.floopy.ai/v1/chat/completions \
-H "Authorization: Bearer $FLOOPY_API_KEY" \
-H "x-floopy-fallback-provider: openai" \
-H "x-floopy-fallback-model: o3-mini" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-reasoner", "messages": [{"role": "user", "content": "Hello"}]}'