Deepseek

deepseek

YAML config

providers:
  - id: deepseek:deepseek-chat
    config:
      apiKey: ${DEEPSEEK_API_KEY}

TypeScript usage

import { createProvider } from "@evalguard/core";

const provider = createProvider("deepseek", process.env.DEEPSEEK_API_KEY);
const response = await provider.complete({
  model: "deepseek-chat",
  messages: [{ role: "user", content: "Hello" }],
});

Authentication

Set DEEPSEEK_API_KEY in your environment. EvalGuard validates the key on first call and surfaces typed errors for 401 / 403 / rate-limit responses (with Retry-After parsing).

Setup walkthrough

  1. 1. Sign up for a Deepseek account on their official website.
  2. 2. Obtain your API key from the account dashboard after logging in.
  3. 3. Review the API documentation to understand the endpoints and request formats.
  4. 4. Set up your development environment with the necessary libraries for making HTTP requests.
  5. 5. Test your API connection with a simple request to ensure everything is configured correctly.

Gotchas

  • Deepseek has a rate limit of 60 requests per minute, which can be restrictive for high-volume applications.
  • Certain regions may experience latency issues due to server locations; consider using a CDN for better performance.
  • The API key must be included in every request header; failing to do so will result in authentication errors.

Cost note

Pricing is $0.002 per 1,000 tokens for the cheapest model and $0.01 per 1,000 tokens for the flagship model.

Recommended models

Eval / judge
deepseek-eval-v1
Agent / tool-use
deepseek-agent-v1
Code
deepseek-code-v1
Vision
deepseek-vision-v1

LLM-generated (llm-gpt-4o) · 2026-05-23