Use DeepSeek in GitHub Copilot

Steps

1. 1
   Step 1.

   Open the Language Models editor: choose Manage Language Models (gear icon) in the chat model picker, or run Chat: Manage Language Models from the Command Palette.

2. 2
   Step 2.

   Select Add Models, then Custom Endpoint. Enter a group name, a display name and your DeepSeek API key, and choose the Chat Completions API type.

3. 3
   Step 3.

   VS Code opens chatLanguageModels.json. Configure the model as below (Custom Endpoint schema from VS Code docs, filled with DeepSeek's endpoint and ids), then save.

   JSON

   Copy

   [
     {
       "name": "DeepSeek",
       "vendor": "customendpoint",
       "apiKey": "${input:deepseekApiKey}",
       "apiType": "chat-completions",
       "models": [
         {
           "id": "deepseek-flash",
           "name": "DeepSeek Flash",
           "url": "https://api.deepseek.com/chat/completions",
           "toolCalling": true,
           "vision": true,
           "contextWindow": 1000000,
           "maxOutputTokens": 384000
         },
         {
           "id": "deepseek-v4-pro",
           "name": "DeepSeek V4 Pro",
           "url": "https://api.deepseek.com/chat/completions",
           "toolCalling": true,
           "vision": false,
           "contextWindow": 1000000,
           "maxOutputTokens": 384000
         }
       ]
     }
   ]

4. 4
   Step 4.

   Select the model from the chat model picker. Restart VS Code if it does not show up.