Model Configuration

Tencent WorkBuddy comes with a wide range of built-in models covering reasoning, multimodal, and image processing capabilities. Custom models can be configured through a visual interface — no need to manually edit configuration files.

Auto Mode

When Auto Mode is enabled, Tencent WorkBuddy automatically selects the most suitable model for each task.

Built-in Models

Tencent WorkBuddy provides built-in model options for everyday tasks, reasoning, and multimodal scenarios. The available model list may vary by product version, account type, and service availability.

Custom Models

• Model list: Configured custom models are displayed in a list, showing the model name and provider icon.
• Add / Edit / Delete: All operations are performed through a dialog — no configuration file editing required.
• Model selector integration: The model selector in the chat interface shows a custom model group and supports quick navigation to the configuration page.

You can manage custom models through the Settings → Model page. Configuration is automatically saved after you click Save. When you select a standard provider, capability flags (tool calling, image input, reasoning mode, etc.) are filled in automatically.

Note: Models previously configured via ~/.codebuddy/models.json continue to work after the UI upgrade and can be viewed, edited, or deleted through the interface — no need to edit the file directly.

Integration Methods

Provider Presets

Select a provider from the list (or choose Custom API). The URL, model list, and capability flags (tool calling, image input, reasoning mode, etc.) are filled in automatically — you only need to enter your API Key to save and start using the model.

• Custom API: Built-in presets for major providers — connect with one click.

Local Deployment (Ollama)

Ollama is an open-source tool for running large models locally. After installation, pull and run a model with a single command. Ollama listens on a local HTTP port (default: 11434) and exposes an OpenAI-compatible interface that Tencent WorkBuddy can connect to directly.

Use cases:

1. Data privacy / on-premises compliance — code and conversation content never leaves your machine; suitable for finance, government, and sensitive projects.
2. Zero cost — no token consumption, no paid API key required; use your own hardware as compute.
3. Offline use — works on planes, air-gapped machines, and environments without internet access.

Custom

If your model service is not in the provider list, select Custom, then manually enter the URL, API Key, and model name.

Custom Protocol

When your model service uses a non-standard URL path (e.g., behind a gateway or proxy), enable the Custom Protocol toggle in advanced settings. Tencent WorkBuddy will then send requests directly to the URL you entered, bypassing path validation and auto-completion.

State	Behavior
Off (default)	Uses the standard /chat/completions path; validates and auto-completes the endpoint URL
On	Sends requests directly to the entered URL; skips path validation and auto-completion