Open source · Python · PyQt6

Your local AI
workspace

A desktop GUI for Ollama that knows your hardware, recommends the right model, and keeps everything private — no cloud, no subscriptions.

LocalDesk
CPUAMD Ryzen 9 · 16 cores · 4800 MHz
RAM22.4 GB available / 32 GB
GPUNVIDIA RTX 4070 · 9.2 GB free
Best fit for your PC
qwen2.5:14b
GPU mode · 9.2 GiB free VRAM · disk OK
qwen2.5:14b local
You
Write a Python function that parses JSON from a file and returns a typed dataclass
qwen2.5:14b
Sure! Here's a typed approach using dataclasses and json:

from dataclasses import dataclass
import json

@dataclass
class Config:
Ask anything...
Send
Everything you need,
nothing you don't

LocalDesk wraps Ollama in a polished desktop experience designed for developers who want full control.

🧠

Hardware-aware recommendations

Scans your CPU, RAM, and NVIDIA VRAM on startup and recommends the largest model that will run comfortably — no guessing.

Real-time streaming

Responses stream token-by-token with a blinking cursor. Incremental UTF-8 decoding means nothing breaks on non-ASCII output.

💬

Conversation memory

Keeps context across the last 12 messages so the model actually remembers what you were just talking about.

📁

File attachments

Drop a .py or .txt file into the chat and it's included inline. Vision models get images passed directly.

💾

Save & restore chats

Every conversation can be saved as JSON and reopened later. Your chat history stays local, always.

🛑

Stop generation

Kill a running response instantly. The partial output is kept so you can see what the model had before you stopped it.


What's built, what's coming

The core is solid. These are next.

✓ Shipped

Chat UI with streaming
Hardware detection
Automatic model recommendation
Model download via ollama pull
Save & load chat history
File & image attachments
macOS support (Apple Silicon & Intel)
Settings tab (system prompt, username)
Temperature control via Ollama API

→ v2 planned

Markdown rendering in chat
System tray / quick assistant mode
RAG support (local documents)
Voice commands
Get LocalDesk
running in minutes

Works on Linux, Windows, and macOS. Requires Ollama to be installed and available in your PATH.

⚠️ Windows Defender may warn — no digital signature yet. You can always run from source.

# 1. Install Ollama
curl -fsSL https://ollama.com/install.sh | sh
 
# 2. Download LocalDesk.bin from Releases
chmod +x LocalDesk.bin
./LocalDesk.bin
Simple, solid stack
Python 3.11+ PyQt6 Ollama psutil subprocess threading MIT License Linux · Windows · macOS