SvelteKit-based WebUI by allozaur · Pull Request #14839 · ggml-org/llama.cpp

Posted by jacek2023@reddit | LocalLLaMA | View on Reddit | 12 comments

"This PR introduces a complete rewrite of the llama.cpp web interface, migrating from a React-based implementation to a modern SvelteKit architecture. The new implementation provides significant improvements in user experience, developer tooling, and feature capabilities while maintaining full compatibility with the llama.cpp server API."

✨ Feature Enhancements

File Handling

Dropdown Upload Menu: Type-specific file selection (Images/Text/PDFs)
Universal Preview System: Full-featured preview dialogs for all supported file types
PDF Dual View: Text extraction + page-by-page image rendering
Enhanced Support: SVG/WEBP→PNG conversion, binary detection, syntax highlighting
Vision Model Awareness: Smart UI adaptation based on model capabilities
Graceful Failure: Proper error handling and user feedback for unsupported file types

Advanced Chat Features

Reasoning Content: Dedicated thinking blocks with streaming support
Conversation Branching: Full tree structure with parent-child relationships
Message Actions: Edit, regenerate, delete with intelligent branch management
Keyboard Shortcuts:
Ctrl+Shift+N: Start new conversation
Ctrl+Shift+D: Delete current conversation
Ctrl+K: Focus search conversations
Ctrl+V: Paste files and content to conversation
Ctrl+B: Toggle sidebar
Enter: Send message
Shift+Enter: New line in message
Smart Paste: Auto-conversion of long text to files with customizable threshold (default 2000 characters)

Server Integration

Slots Monitoring: Real-time server resource tracking during generation
Context Management: Advanced context error handling and recovery
Server Status: Comprehensive server state monitoring
API Integration: Full reasoning_content and slots endpoint support

🎨 User Experience Improvements

Interface Design

Modern UI Components: Consistent design system with ShadCN components
Responsive Layout: Adaptive sidebar and mobile-friendly design
Theme System: Seamless auto/light/dark mode switching
Visual Hierarchy: Clear information architecture and content organization

Interaction Patterns

Keyboard Navigation: Complete keyboard accessibility with shortcuts
Drag & Drop: Intuitive file upload with visual feedback
Smart Defaults: Context-aware UI behavior and intelligent defaults (sidebar auto-management, conversation naming)
Progressive Disclosure: Advanced features available without cluttering basic interface

Feedback & Communication

Loading States: Clear progress indicators during operations
Error Handling: User-friendly error messages with recovery suggestions
Status Indicators: Real-time server status and resource monitoring
Confirmation Dialogs: Prevent accidental data loss with confirmation prompts

[-]

Dry_Management_8203@reddit

Okay, this looks great but, how does a beginner like me actually go about downloading, installing, and using it.

A step by step guide would be extremely helpful.

yc22ovmanicom@reddit

Just download the new version. It's terrible; all the pretty themes are gone, only two awful ones remain. The think process isn't displayed, even if I check the box to show the think process; only the final result. PP generation speed isn't displayed. There's a button to edit old messages, but it doesn't work properly and messes up the entire conversation history. Hotkeys only work with the English keyboard layout. The only positive is that it displays the current context size. If this were the default, it would be the end of the llama.cpp webui.

MatterMean5176@reddit

Growing pains bruv. But bring back my aqua.. please.. ;)

Heya, I'm taking it for a whirl now. Exciting changes. No reasoning display for me either like the other guy said. Cheers.

Secure_Reflection409@reddit

It looks beautiful but I think it just crashed when I used it with Magistral.

Double_Cause4609@reddit

Awesome, new UI. Tools, please. Tools please.
Cool, file handling. Yes, Yes, yes! Tools please.
Advanced Chat features? Awesome. Like, tools?
Okay, maybe not. Maybe they're somewhere else?
Server integration? Awesome, it'd make sense to throw custom tools in there.
Alright, fair enough. But tools are a really useful feature. Maybe they're somewhere else?
Interface design. Weird, cool features, I guess tools could be here?
No, okay...
Interaction patterns. Surely tools are in here!
No...
Feedback...
Awwww...

Tbh all I want is a simple UI that lets me define custom tools.

legit_split_@reddit

Looks very clean :)

Distinct-Rain-2360@reddit

Regarding the keyboard shortcuts, some are already used by browsers, :

- ctrl-shift-n opens a private window in chromium browsers

- ctrl-shift-d adds a bookmark in firefox browsers and prompts to add all tabs to bookmarks in chrome browsers.

- ctrl-k focuses the search engine bar in firefox browsers

- ctrl-b toggles the bookmark sidebar in firefox browsers

Regarding smart paste, that's something that I disabled in open-webui, it can mess up the html quite badly if you paste markdown with code blocks regularly, please consider keeping it off by default, and you just live with the ugly html because the model has already started prompt processing or thinking and you don't want to reformat.

Regarding smart defaults, please do not use the model to auto title the conversation.

Everything else looks great, its going to be good to be able to just drop a model from a thumb drive or phone into a cloud user's computer, have them download llamacpp from github release and they're all set.

jacek2023@reddit (OP)

I think you should comment in the PR :)

seems like it's already merged, I don't think they'll roll it back or delay it for my nitpicks.

But you can post your comments for the future work

igorwarzocha@reddit

YES PLEASE. I've been on a hunt for a no-bs frontend for Llama.cpp for ages! These fancy apps are doing my head in.

MCP support, /compact, pretty please :)