https://github.com/kashif-e/kollama
A UI client for Ollama written in Compose Multiplatform focused on running Deepseek r1 locally
https://github.com/kashif-e/kollama
compose compose-multiplatform deepseek-chat deepseek-r1 kmp koin kotlin kotlin-multiplatform mistral ollama-app ollama-client ollama-gui ollama-ui
Last synced: 23 days ago
JSON representation
A UI client for Ollama written in Compose Multiplatform focused on running Deepseek r1 locally
- Host: GitHub
- URL: https://github.com/kashif-e/kollama
- Owner: Kashif-E
- License: apache-2.0
- Created: 2025-01-28T21:56:10.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-02-11T10:31:28.000Z (3 months ago)
- Last Synced: 2025-04-02T04:08:34.751Z (about 1 month ago)
- Topics: compose, compose-multiplatform, deepseek-chat, deepseek-r1, kmp, koin, kotlin, kotlin-multiplatform, mistral, ollama-app, ollama-client, ollama-gui, ollama-ui
- Language: Kotlin
- Homepage:
- Size: 2.16 MB
- Stars: 98
- Watchers: 1
- Forks: 9
- Open Issues: 1
-
Metadata Files:
- Readme: README.MD
- License: LICENSE
Awesome Lists containing this project
README
# Kollama
Kollama is a modern, cross-platform chat interface for Ollama models built with Kotlin Multiplatform and Compose Multiplatform. It provides a sleek, user-friendly way to interact with local AI models while offering advanced features like code formatting, markdown support, and animated UI elements.
## Features
- 🚀 **Cross-Platform Support**: Built with Kotlin Multiplatform and Compose Multiplatform
- 💬 **Real-time Chat Interface**:
- Streaming responses with real-time updates
- Concurrent chat support with efficient job management
- Smart session handling with LRU cache
- 🎨 **Rich Markdown Support**:
- Syntax-highlighted code blocks with customizable themes
- Tables and complex formatting
- Custom typography system
- Optimized rendering for large messages
- 🔄 **Advanced Message Handling**:
- Real-time message streaming with state management
- Robust error handling and retry mechanism
- Optimized clipboard operations
- Sophisticated message state tracking
- 💾 **Persistent Storage**:
- SQLDelight for efficient, type-safe SQL operations
- Cross-platform database support
- Optimized query performance
- 🎯 **Advanced Model Integration**:
- Dynamic model switching with state preservation
- Model details display with animated marquee
- Real-time model status indicators
- 🎠**Polished UI/UX**:
- Animated right sheet with gesture support
- Custom animated borders and transitions
- Fluid dark/light theme switching
- Responsive layout with efficient recomposition
- Pulsating status indicators and loading states
## Screenshots
## Prerequisites
- JDK 17 or higher
- Ollama installed and running (default: localhost:11434)
- At least one Ollama model downloaded
## Getting Started
1. Install Ollama from [Ollama's official website](https://ollama.ai)
2. Pull a model using Ollama CLI:
```bash
ollama pull deepseek-coder:33b
```
3. Clone this repository:
```bash
git clone https://github.com/kashif-e/kollama.git
```
4. Run using Gradle:
```bash
./gradlew run
```
## Architecture
Kollama follows clean architecture principles with a focus on modularity and testability:
### Core Components
- **Presentation Layer**:
- `ChatViewModel`: Manages UI state and user interactions
- `ChatScreen`: Main UI composition
- Custom composables for reusable UI elements
- Advanced state management with Kotlin Flow
- **Domain Layer**:
- Pure Kotlin models for type safety
- Repository interfaces for data abstraction
- Clear separation of concerns
- **Data Layer**:
- `ChatRepository`: Implementation of data operations
- `OllamaService`: Handles AI model communication
- `DatabaseWrapper`: SQLDelight database operations
- Efficient data mapping and transformation
- **Infrastructure**:
- Koin for dependency injection
- SQLDelight for database operations
- Ktor for network communications
### Key Technical Features
#### State Management
```kotlin
data class ChatUiState(
val isLoading: Boolean = false,
val error: String? = null,
val currentSession: ChatSessionDomainModel? = null,
val chatSessions: List = emptyList(),
val selectedModel: OllamaModel? = null,
val availableModels: List = emptyList(),
val thinkingMessage: ChatMessageDomainModel? = null,
val showModelList: Boolean = false,
)
```
#### Efficient Job Management
```kotlin
class ChatJobManager {
private val mutex = Mutex()
private val jobMap: MutableMap = mutableMapOf()
private val lruList = DoublyLinkedList()
// Manages concurrent chat sessions with LRU caching
}
```
#### Custom UI Components
```kotlin
@Composable
fun AnimatedBorderBox(
borderColors: List,
backgroundColor: Color,
shape: Shape = RectangleShape,
borderWidth: Dp = 1.dp,
animationDurationInMillis: Int = 3000,
content: @Composable BoxScope.() -> Unit
)
```
## Project Structure
```
kollama/
├── data/
│ ├── local/
│ │ ├── database/
│ │ └── service/
│ ├── model/
│ └── repository/
├── domain/
│ ├── interfaces/
│ └── model/
├── presentation/
│ ├── components/
│ ├── modifier/
│ ├── state/
│ └── theme/
└── di/
```
## Advanced Features
### Custom Animations
- Animated message borders with gradient effects
- Pulsating status indicators
- Smooth sheet transitions with gesture support
- Marquee animations for model details
### Performance Optimizations
- Efficient recomposition with remember/derivedStateOf
- Smart job management for concurrent chats
- Optimized database operations
- Proper coroutine scope management
### Network & Error Handling
- Robust error handling with custom error types
- Retry mechanisms for failed requests
- Proper timeout and cancellation handling
- Stream processing for real-time updates
## Contributing
1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Push to the branch
5. Create a Pull Request
## Acknowledgments
- [Kotlin Multiplatform](https://kotlinlang.org/docs/multiplatform.html)
- [Compose Multiplatform](https://www.jetbrains.com/lp/compose-multiplatform/)
- [Ollama](https://ollama.ai)
- [SQLDelight](https://cashapp.github.io/sqldelight/)
- [Koin](https://insert-koin.io)