Here it is, the first feature release for DEVONthink 4. Choose from a much wider range of server-based AI models via OpenRouter or by connecting to just any OpenAI-compatible provider. Integrate models from, e.g., Amazon, Meta, Microsoft, xAI, Baidu, Moonshot AI, Qwen, and more. In addition, DEVONthink Europa adds native support for Claude 4.1 Opus, Gemini 2.5 Flash Lite, OpenAI GPT-5, Google Imagen 4, and the latest models by Mistral. Connect to your institution's Ollama installation with a new configuration option.

To further automate your work, describe your task to the the AI-powered script assistant. It generates AppleScript and JavaScript for Automation (JXA) scripts and even corrects common scripting issues. Edit images, both generated or dropped in from other sources, with prompts using common AI image generation models. As generative AI is notoriously bad at maths, the Chat and other AI-assisted functions now support mathematical functions, and if you're working in academia, retrieval-augmented generation (RAG) gets direct access to the huge document collection on arXiv. In addition, DEVONthink Europa comes with a new Quick Look extension for previewing Spotlight results on macOS Sequoia and later, and more responsive editing of especially larger Markdown documents. It also updates and sorts lists and empties the trash significantly faster.

This maintenance release introduces support for Google Imagen 4 (Preview), along with several improvements and fixes. We enhanced handling of relative links in Markdown documents to allow spaces, and we corrected an issue with list indentation. We also improved document language detection, which can influence Chat-suggested properties like tags. The Favorites section, e.g., in the Move To context menu now correctly lists all favorited groups. Colored groups also show their designated colors in more locations, such as the Path bar. For AI features, we expanded the prompt field in the Chat and Help assistants to accommodate longer prompts. We extended the timeout period for image generation, and queries will retry up to two times when necessary. As always, we focused on increasing the application's stability and reliability.

Our fourth major release — DEVONthink 4.0 Copernicus — brings a host of powerful new features and improvements. DEVONthink now supports external AI providers, both commercial and local AI, as a complement to its own internal engine and implemented with privacy and data safety in mind. You can also write search queries in natural language, let the AI summarize or transform text, and extract text from images or handwritten notes. Use speech-to-text to transcribe and save the speech from media files. You can even create images using AI, directly within DEVONthink.

Store important legal or financial documents in audit-proof databases and as PDF/A. Text in PDFs without text layer and in images is searchable without OCR, and versioning lets you return to previous versions of a document. Edit a document's table of contents, adjust spacings, and fine-tune the WYSIWYG display for Markdown. Organize your favorites, smart groups, and smart rules into groups. We also added new inspectors, most notably a graph that visualizes document relationships. The rewritten web interface of the Server edition is built on a modern foundation, more powerful yet easier to use.

For automators of any skill level, there are new and improved smart actions, including some utilizing AI, as well as many new AppleScript commands. Batch processing now lets you save and reuse configurations.

This release brings support for cutting-edge AI models such as Anthropic's Claude 4, Sonnet, and Opus, Perplexity's Sonar Reasoning models, and Mistral Medium 3. Icons make it easier to see at a glance which AI provider you're working with, separate AI > Summarization settings let you better control how DEVONthink summarizes documents, and you can exclude documents or whole groups from being ever accessed by AI. The Chat assistant can draft an email for you, and it has more reliable rendering of lists, code blocks, and MathJax. Finally, there have been numerous under-the-hood changes made to reduce costs, control the exchange of information, and filter irrelevant information to produce better responses.

AI aside, a new toolbar button simplifies creating compressed archives of databases, while improvements in metadata handling, Markdown editing, and PDF processing enhance usability and performance. In addition, DEVONthink sends us some anonymized data that can help us determine, e.g., if some features or actions are more popular than others. Of course, you can switch this off in the app's settings. The update also includes numerous fixes addressing stability issues, UI inconsistencies, and script compatibility.

Today's release adds OpenAI's new image generation engine and Google Gemini 2.5. We also replaced some AI models with newly improved ones, like GPT 4.1. You can now adjust the size of the font used in your interactions with AI, e.g, the Chat assistant. Other improvements include better transcriptions and timestamps, better link handling in the assistant, and more useful resolutions for placeholders like prices or dates. A new AI-based script tries to geolocate images in your databases.

For faster searching, there are new keyboard shortcuts for changing the search scope. Open clicked links in foreground windows, including external links, or hold the Command key to temporarily toggle between opening the document externally or in DEVONthink. If you use a Doxie scanner, DEVONthink recognizes incoming scans for OCR. We've also added some new external scripts to use in smart rules and batch processing.

On the interface side, we have harmonized text sizes across certain popovers and made other refinements to the interface, including refreshing icons and fine-tuning spacing. For scripters, this release shows a significant performance improvement when making filtered queries. And finally, we have made migrating resources from version 3 more reliable, and fixed some bugs and crashes as we work to make version 4 even more stable and performant.

This is the first public beta of our next generation of DEVONthink, bringing powerful new AI capabilities and significant improvements across the board.

Interact with your chosen AI engine through our new Chat assistant, accessible via a convenient inspector or popover. Ask questions about your documents, modify properties, conduct web searches, all through natural language. Let AI translate your requests into DEVONthink's native search syntax or summarize and transform text like Apple's Writing Tools. Detect text within images and handwritten notes, transcribe audio/video, and even generate images directly in DEVONthink.

PDFs lacking OCR are now automatically searchable. For sensitive documents, legal, financial, etc., create audit-proof databases and convert PDFs to PDF/A for storage. We've greatly improved document editing, including options to modify the table of contents and create bookmarks within PDFs. DEVONthink now retains multiple versions of edited documents allowing quickly going back to previous versions. New AI-powered inspectors provide intelligent tagging support and visual relationship graphs.

Team collaboration is smoother than ever with our completely rebuilt web sharing interface. Simpler yet more powerful.

For automation enthusiasts, we've incorporated new and improved smart actions, including AI-powered options for users of all skill levels. The modernized scripting dictionary offers expanded commands, properties, and parameters, while batch processing allows you to save workflows as reusable configurations.

We've continued our commitment to usability, performance and reliability throughout the app.