llama.cpp Web UI

Table of Contents

Features

Chat Interface

File Attachments

Conversation Management

Advanced Rendering

Multi-Model Support (ROUTER mode)

Keyboard Shortcuts

Developer Experience

Getting Started

Prerequisites

1. Install Dependencies

2. Start llama-server

3. Start Development Servers

Development Workflow

Tech Stack

Key Dependencies

Build Pipeline

Development Build

Production Build

SvelteKit Configuration

Integration with llama-server

Architecture

High-Level Architecture

Layer Breakdown

Routes (`src/routes/`)

Components (`src/lib/components/`)

Hooks (`src/lib/hooks/`)

Stores (`src/lib/stores/`)

Services (`src/lib/services/`)

Data Flows

MODEL Mode (Single Model)

ROUTER Mode (Multi-Model)

Detailed Flow Diagrams

Architectural Patterns

1. Reactive State with Svelte 5 Runes

2. Unidirectional Data Flow

3. Per-Conversation State

4. Message Branching with Tree Structure

5. Layered Service Architecture

6. Server Role Abstraction

7. Modality Validation

8. Persistent Storage Strategy

Testing

Test Types

Running Tests

Storybook Development

Linting and Formatting

Project Structure

Related Documentation