On September 22, 2026, DeepL launched "DeepL MCP," allowing users to access its translation capabilities directly from AI assistants such as Microsoft Copilot, ChatGPT, and Claude via the Model Context Protocol (MCP). This eliminates the need to copy and paste text or documents created in an AI assistant into a separate translation service, enabling direct translation right from the AI chat interface.
PLUS ULTRAProduct LaunchesDeepLDeepL MCP
DeepL Supports MCP, Enabling Direct Translation from ChatGPT and Copilot
PLUS ULTRA by Amenoyomi
This article is a translation. Read the Japanese original
With DeepL MCP, individual messages or multiple texts can be translated into 1 or more languages. For document translation, it can process files while maintaining formatting and layouts, including tables, headings, and images. Additionally, since DeepL's glossaries and custom rules can be applied, users can reflect company-specific terminology and tone in translations performed within AI assistants.
In terms of management, organization administrators can control access rights at will, and in Microsoft Teams, DeepL can be set as the default translation provider. Regarding usage fees, it is based on the user's existing DeepL subscription, and DeepL has stated that transmitted text will not be used for DeepL model training.
Support is available through the Microsoft Marketplace, Agent Store, and connector directories, allowing integration into existing products and tools.
PLUS ULTRAby Amenoyomi
The traditional practice of switching between tools for every translation is shifting toward a system where the entire process is completed within the AI assistant.
Beyond simply eliminating the hassle of copying and pasting, users can move from drafting to reviewing and multilingual adjustments within a single conversational flow, preventing interruptions and optimizing workflows.
Furthermore, because DeepL's glossaries and custom rules can be applied directly, there is a practical advantage: even when translating via an AI assistant, company-specific technical terms and tones can be maintained.
At its core, this change is not merely about enhancing translation functions, but rather a structural shift where AI becomes the "operational hub" that freely manipulates external specialized services. By using the AI assistant as the starting point to link existing services such as translation and search, the workflow will evolve from switching between tools to an integrated, AI-centric process.
Sources
- ChatGPTやCopilotから直接翻訳 「DeepL」のMCP対応で多言語業務はどう変わる？ （ITmedia AI+、2026-09-25）