DeepSeek Documents GitHub Copilot CLI BYOK Setup

Migrated with permission from DeepSeek V4 Pro, a related editorial property operated by our team.

Checked on June 24, 2026: DeepSeek's official GitHub Copilot CLI page says teams should use the Anthropic-compatible endpoint, warns that the OpenAI provider triggers a 400 reasoning-content error, sets explicit token ceilings for deepseek-v4-pro, and clarifies that offline mode still sends prompts to DeepSeek.

What we verified on June 24, 2026

DeepSeek now has an official GitHub Copilot CLI integration page inside the agent-integrations section of the API docs.

The page publishes a strict provider rule: use anthropic as the provider type and point Copilot CLI to https://api.deepseek.com/anthropic.

DeepSeek documents the actual failure mode behind a common setup error: the openai provider type triggers a 400 error that says The reasoning_content in the thinking mode must be passed back to the API. because Copilot CLI's OpenAI integration does not echo reasoning_content back on later requests.

The official page also publishes explicit token ceilings for the custom DeepSeek route: 840000 max prompt tokens and 128000 max output tokens for deepseek-v4-pro.

DeepSeek adds an operational boundary around offline mode: COPILOT_OFFLINE=true only blocks GitHub API calls, while prompts still go to api.deepseek.com.