deepseek_api_chat

DeepSeek Hybrid Agent (GUI) is a Python + PyQt6 desktop chat client designed for stable LLM streaming under VPN conditions (ProtonVPN/OpenVPN). It automatically detects VPN/proxy settings, supports SOCKS5, and includes adaptive timeouts, smart retry with exponential backoff, and a circuit breaker to prevent repeated network failures.

Deepseek Api Chat

deepseek_chat.py is a PyQt6-based hybrid LLM chat client optimized for VPN environments, featuring automatic VPN/proxy detection, real-time streaming, smart anti-lag storage, hybrid DeepSeek API / Ollama execution, and automatic code completion stopping. 🚀

Getting Started

This section describes the recommended setup flow to run the application locally.

1. Clone the repository

git clone https://github.com/Qatwi/deepseek_api_chat.git
cd  deepseek_api_chat

2. (Optional but recommended) Create a virtual environment

python3 -m venv .venv
source .venv/bin/activate

This isolates dependencies and avoids polluting the system Python environment.

3. Install dependencies

pip install -r requirements.txt

If you do not use requirements.txt, you can install manually:

pip install pyqt6 requests pysocks cryptography
pip install "requests[socks]"

4. (Optional) Install Ollama for local inference

Ollama is not required, but recommended as a fallback backend.

curl -fsSL https://ollama.com/install.sh | sh

After installation, pull a model:

ollama pull claude-coder

5. Run the application

python3 deepsek_chat.py

On first launch:

• VPN and proxy configuration is auto-detected
• A configuration file is created automatically
• You will be prompted to enter a Claude API key (or select Ollama-only mode)

No manual configuration is required.

Notes

• The application does not require environment variables
• Configuration is generated and managed automatically
• Local configuration files are intentionally ignored by Git