How to Connect DeepSeek Models in VS Code Using Continue and SiliconFlow

Originally I wanted to integrate DeepSeek in IDEA, but the official API platform left me stunned.

Unfortunately, I couldn’t recharge, so the article ends here… but wait, there is a solution—follow the steps below.

Register a SiliconFlow account

Visit https://www.siliconflow.com/ and click Register to receive 20 million tokens.

Register with a phone number and verification code, optionally using the invitation code h3pcklak or the invitation link https://cloud.siliconflow.cn/i/h3pcklak .

After logging in you will see a huge selection of models, including Huawei‑cloud‑deployed full‑blood R1 and V3.

In the left‑hand API Keys menu generate an API key.

Integrate DeepSeek via the Continue plugin

Continue plugin overview

Website: https://www.continue.dev/

Continue is an open‑source AI code‑assistant plugin for VS Code and JetBrains IDEs that provides real‑time code completion suggestions and can connect to various language models.

Install the Continue plugin

Search for “Continue” in the IDE’s plugin marketplace and install it.

After installation click the Continue icon on the right sidebar and then the settings button to open the configuration file.

Configuration

Copy the JSON configuration below, replace the apiKey fields with your own SiliconFlow API keys, and save the file.

{
  // Additional configuration for R1, V3, V2.5 models
  "models": [
    {
      "model": "claude-3-5-sonnet-latest",
      "provider": "anthropic",
      "apiKey": "",
      "title": "Claude 3.5 Sonnet"
    },
    {
      "title": "deepseek-ai/DeepSeek-R1",
      "model": "deepseek-ai/DeepSeek-R1",
      "contextLength": 30000,
      "provider": "openai",
      "apiBase": "https://api.siliconflow.cn/v1",
      "apiKey": "输入自己的API keys",
      "requestOptions": {"extraBodyProperties": {"transforms": []}}
    },
    {
      "title": "deepseek-ai/DeepSeek-V3",
      "model": "deepseek-ai/DeepSeek-V3",
      "contextLength": 30000,
      "provider": "openai",
      "apiBase": "https://api.siliconflow.cn/v1",
      "apiKey": "输入自己的API keys",
      "requestOptions": {"extraBodyProperties": {"transforms": []}}
    },
    {
      "title": "deepseek-ai/DeepSeek-V2.5",
      "provider": "openai",
      "model": "deepseek-ai/DeepSeek-V2.5",
      "contextLength": 30000,
      "apiBase": "https://api.siliconflow.cn/v1",
      "apiKey": "输入自己的API keys",
      "useLegacyCompletionsEndpoint": false
    }
  ],
  "tabAutocompleteModel": [
    {
      "title": "deepseek-ai/DeepSeek-V2.5",
      "provider": "openai",
      "model": "deepseek-ai/DeepSeek-V2.5",
      "contextLength": 30000,
      "apiBase": "https://api.siliconflow.cn/v1",
      "apiKey": "输入自己的API keys",
      "useLegacyCompletionsEndpoint": false
    }
  ],
  "tabAutocompleteOptions": {
    "template": "Please teach me what I should write in the `hole` tag, but without any further explanation and code backticks, i.e., as if you are directly outputting to a code editor. It can be codes or comments or strings. Don't provide existing & repetitive codes. If the provided prefix and suffix contain incomplete code and statement, your response should be able to be directly concatenated to the provided prefix and suffix. Also note that I may tell you what I'd like to write inside comments. 
{{{prefix}}}<hole></hole>{{{suffix}}}

Please be aware of the environment the hole is placed, e.g., inside strings or comments or code blocks, and please don't wrap your response in ```. You should always provide non-empty output.
",
    "maxPromptTokens": 2048,
    "prefixPercentage": 0.85,
    "maxSuffixPercentage": 0.15,
    "debounceDelay": 500,
    "multilineCompletions": "always",
    "slidingWindowPrefixPercentage": 0.75,
    "slidingWindowSize": 350,
    "maxSnippetPercentage": 0.6,
    "recentlyEditedSimilarityThreshold": 0.3,
    "useCache": true,
    "onlyMyCode": false,
    "useOtherFiles": false,
    "useRecentlyEdited": true,
    "recentLinePrefixMatchMinLength": 7
  },
  "customCommands": [
    {
      "name": "test",
      "prompt": "{{{ input }}}

Write a comprehensive set of unit tests for the selected code. It should setup, run tests that check for correctness including important edge cases, and teardown. Ensure that the tests are complete and sophisticated. Give the tests just as chat output, don't edit any file.",
      "description": "Write unit tests for highlighted code"
    }
  ],
  "contextProviders": [
    {"name": "diff", "params": {}},
    {"name": "folder", "params": {}},
    {"name": "codebase", "params": {}}
  ],
  "slashCommands": [
    {"name": "share", "description": "Export the current chat session to markdown"},
    {"name": "commit", "description": "Generate a git commit message"}
  ]
}

After saving, the added models appear in the chat window.

Test the setup; note that DeepSeek models on SiliconFlow are currently very popular, and using R1 may result in slower responses.