Model Selection
Introduction
The plugin currently supports two models for you to choose from: fast and deep. The Express model excels at core feature implementation, code refactoring, and daily development tasks, offering ultimate inference speed and lightning-fast responses. The deep model specializes in architecture design and core feature implementation, providing deep reasoning capabilities. Please choose the appropriate model based on your task requirements to proceed with subsequent tasks. Please note that this model is only available to individual Pro subscribers.
Model Selection
- In Smart Guide, AI Toolbox, and Smart Chat, you can select either "fast" or "deep" based on your specific task scenario. Additionally, you can choose whether to enable "Thinking Mode".
Thinking Mode Overview:
| Dimension | Low Thinking Mode | Medium Thinking Mode | High Thinking Mode |
|---|---|---|---|
| Analogy | "Gut reaction": Like casual chat, instant and spontaneous. | "Standard problem-solving": Like doing homework, step-by-step but not overly complex. | "Expert panel discussion": Like scientific research, involving repeated reasoning and self-correction. |
| Speed | Ultra-fast (milliseconds) | Moderate (seconds) | Slower (may take tens of seconds or longer) |
| Use Cases | Casual chat, translation, simple Q&A, creative brainstorming | Writing emails, summarizing articles, routine coding, general logic puzzles | Complex math, advanced code architecture, legal analysis, scientific reasoning |
| Pros | Rapid response, low cost, smooth and natural flow | Balances speed and accuracy, highly cost-effective | Extremely high accuracy, handles multi-step complex logic, reduces hallucinations |
| Cons | Prone to errors, struggles with complex logical traps | May lack rigorousness when handling extremely complex issues | Time-consuming, high computational resource consumption |
- The model will generate responses based on the mode you have selected.
