Online Model Settings
Online Model Settings
WordOllama supports integration with ChatGPT and various ChatGPT-compatible model APIs.
Model Name
Enter the name used to invoke the model.
Common Mistakes:
| Model Series | Incorrect Name | Correct Name |
|---|---|---|
| ChatGPT | ChatGPT/OpenAI | gpt-3.5-turbo/gpt-4o/gpt-4o-mini/… |
| Tongyi Qwen | Tongyi/Qianwen/tongyi | qwen-long/qwen-max/qwen-plus/… |
| Zhipu GLM | glm/Zhipu/glm4 | glm-4-flash/glm-4/glm-4-plus/… |
| Doubao | Doubao/doubao/Doubao-pro-32k | Check the backend for the exact name, e.g., ep-20241022214223-k2g4z (as of October 30, 2024) |
API Endpoint
Specify the API Endpoint address for the model.
ChatGPT Endpoint
- Official:
https://api.openai.com/v1/chat/completions
Endpoints for Some Domestic Models
| Model | Endpoint |
|---|---|
| DeepSeek | https://api.deepseek.cn/v1/chat/completions |
| Tongyi Qwen | https://dashscope.aliyuncs.com/compatible-mode/v1/chat/completions |
| Zhipu GLM | https://open.bigmodel.cn/api/paas/v4/chat/completions |
| Doubao | https://ark.cn-beijing.volces.com/api/v3/chat/completions |
| KIMI | https://api.moonshot.cn/v1/chat/completions |
API Key
Enter a valid API Key obtained from the model provider.
Tokens Limit
Controls the output length. Refer to Tokens Limit (see Local Model Settings) for details.
Temperature Settings
Controls output randomness. Refer to Temperature Settings (see Local Model Settings) for details.
Model Management
Add Model
- Click the “Add Model” button.
- In the popup, enter:
- Model Name (official standard name)
- Endpoint
- API Key
- Click “Confirm” to save.
Load Model
- Select the target model from the list (click any cell).
- Click the “Load Model” button.
- The system will auto-fill the related configuration.
Delete Model
- Select the model to delete from the list (click any cell).
- Click the “Delete Model” button.
- Confirm the deletion.