Replies: 2 comments
-
|
No |
Beta Was this translation helpful? Give feedback.
0 replies
-
|
Why are you line up ? |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
🌟 What's New in v1.4.0? 🌟
🎂 Dify turns 2! New look, new brand, same mission.
Imagine if every idea could become an AI agent. With Dify, ‘if’ you can ask it, you can build it.
Check out the new Dify.AI and start building your if today.
🚀 New Features & Enhancements
🌙 Dark Mode
We’ve rolled out a theme switcher that allows you to easily toggle between light and dark modes, perfect for those late-night coding sessions. Click your avatar to experience it! This feature helps reduce eye strain and allows you to customize your user interface to match your preference. Thanks to @JzoNgKVO in PR #18093 for making this possible.
🖼️ Multi-Modal LLM Output
Our platform now supports multimodal outputs, allowing LLMs to handle both text and images within the LLMNode. This feature enhances the richness of data interaction and boosts your model's capabilities, courtesy of @QuantumGhost's efforts in PR #17372. You can now try these new features using the Gemini 2.0 Flash Exp model! This integration ensures that your AI can process and respond with both text and visual data seamlessly.
✏️ Editable Questions
We’ve added greater control over your questions, allowing them to be edited before they go live. This update ensures that you can tweak questions for clarity and relevance, as implemented by @HyaCiovo in PR #19117.
🔧 Bug Fixes and Improvements
Enhanced Redis Management: Support Redis client-side caching for optimized data handling (though it’s disabled out-of-the-box), improving overall system performance. Credits to @bowenliang123 in PRs feat: enable Redis client-side caching #19493 and chore: disable redis client-side caching by default #19524.
Robust Workflow Management: Addressed numerous inconsistencies and bugs in workflows, ensuring smoother, more reliable operations across the board. Our workflows can now handle node failures more gracefully, thanks to contributions from @Kevin9703 in PR fix(workflow_cycle_manage): failed nodes were not updated in workflow_node_executions #18994.
These features and improvements reflect the feedback from our amazing community and aim to elevate both usability and functionality. We hope these updates significantly enhance your user experience!
Big thanks to all contributors for steering this version forward. Happy experimenting! 🙌
Upgrade Guide
Docker Compose Deployments
Back up your customized docker-compose YAML file (optional)
Get the latest code from the main branch
Stop the service. Please execute in the docker directory
Back up data
tar -cvf volumes-$(date +%s).tgz volumesUpgrade services
Source Code Deployments
Stop the API server, Worker, and Web frontend Server.
Get the latest code from the release branch:
Update Python dependencies:
cd api uv syncThen, let's run the migration script:
Finally, run the API server, Worker, and Web frontend Server again.
What's Changed
json_repairto fix invalid json while generate structured output by @ZombieBlue in feat(api): usejson_repairto fix invalid json while generate structured output #18977.editorconfigby @QuantumGhost in Enhance Code Consistency Across Repository with.editorconfig#19023/siteAPI by @lexmin0412 in Add/siteAPI #19631New Contributors
/siteAPI #19631Full Changelog: 1.3.1...1.4.0
This discussion was created from the release v1.4.0.
Beta Was this translation helpful? Give feedback.
All reactions