Models
You can add custom models for CodeArts Agent on the console and the client, with some differences between the two. For details, see Table 1.
| Operation Entry | Applicable Package Edition | Permission | Model Application Scope |
|---|---|---|---|
| Console | Professional edition | Enterprise administrator |
|
| Client | All | All enterprise members | User-level: available only to the creator. |
This section describes how enterprise members add custom models on the client. For information on adding custom models on the console, see Configure Enterprise Custom Models.
Constraints
| Category | Description |
|---|---|
| API format | Only the OpenAI Chat Completions format and Anthropic Messages format are supported. |
| Calling tool | Custom models cannot be invoked in the Ask mode. They can be invoked only by built-in system agents, AgentTeam, and custom agents. |
| Token consumption | When you call custom models, tokens from the CodeArts Agent package are not consumed. However, to minimize inference latency, the code generation function uses dedicated built-in models and does not support custom models. |
| Uniqueness | The provider and model ID combination of each model you add must be unique. |
Prerequisites
- Your account has been added as an enterprise member.
- If you are using CodeArts Agent professional edition, ensure that the enterprise administrator has enabled model customization for members on the console. For details, see Configuring Custom Models.
Viewing Built-in Models
- Log in to CodeArts Agent IDE as instructed in Quick Start.
- Click
in the upper right corner of the page to access the global settings page of the IDE. - In the navigation pane of the Settings page, choose Model.
- View the built-in models in the Built-in area.
If you are using CodeArts Agent professional edition, custom models created by the enterprise administrator on the console are also displayed in addition to the built-in system models. The provider is CodeArts for built-in models and Built-in for enterprises for custom models created on the console.
The page displays all built-in system models and custom models created on the console, regardless of whether they are enabled on the console.
Adding Custom Models
CodeArts Agent allows you to add custom models by specifying the model provider or using custom configurations.
- Via providers: CodeArts Agent is preset with some LLMs from mainstream providers. Once you select the provider and model ID, CodeArts Agent automatically populates the model URL. You only need to configure the API key for a quick setup.
- Via custom configuration: You must manually specify the model URL, model ID, and API key. This method is suitable for integrating self-hosted local models or models from providers not included in the preset list.
- Log in to CodeArts Agent IDE as instructed in Quick Start.
- Click
in the upper right corner of the page to access the global settings page of the IDE. - In the navigation pane of the Settings page, choose Model.
- Click Add Model.
- Select Model Provider, and set parameters by referring to Table 3. Figure 1 Adding a model by specifying the model provider
Table 3 Adding a model by specifying the model provider Parameter
Description
Model Provider
Select the provider to which the model to be added belongs from the drop-down list.
Model ID
Select the ID of the model to add from the drop-down list. The model ID cannot be modified after the model is added. If the required model ID is not in the list, select Custom.
Custom Model ID
This parameter is available only when Model ID is set to Custom.
Enter the model ID obtained from the model provider. Its value contains a maximum of 64 characters. The model ID cannot be modified after the model is added.
Model Name
Enter the name of a model displayed in CodeArts Agent. The name contains a maximum of 64 characters.
API Key
Enter the API key obtained from the model provider.
Custom Model URL
Optional. After you select a provider, the system automatically populates the model URL. To modify the model URL, enable Custom Edit and enter the new model URL. Enter only the base URL of the API. For example, if the model address is https://open.bigmodel.cn/api/paas/v4/chat/completions, enter https://open.bigmodel.cn/api/paas/v4 here.
Context Window Tokens
Optional. Set the input tokens and output tokens.
- Input: indicates the maximum number of tokens that can be received by the model in a single request.
- Output: indicates the maximum number of tokens that can be delivered by the model in a single response.
Description
Optional. Enter a maximum of 200 characters to describe the custom model.
- Click Add Model. The system automatically performs a model connectivity test. A real request will be initiated during the test, consuming a small number of model tokens.
- If the test fails, the page displays possible causes. Modify the model parameters based on the information provided, and then click Test and Add to run the connectivity test again. If you are sure that the model parameters are correct, you can also click Save Directly to finish the model addition.
- If the test passes, the new model is displayed in the Custom model list and is enabled by default.
- Log in to CodeArts Agent IDE as instructed in Quick Start.
- Click
in the upper right corner of the page to access the global settings page of the IDE. - In the navigation pane of the Settings page, choose Model.
- Click Add Model.
- Select Custom Configuration, and set parameters by referring to Table 4. Figure 2 Adding a model through custom configuration
Table 4 Adding a model through custom configuration Parameter
Description
API Format
Select the API format used by the model from the drop-down list. The options include OpenAI Chat Completions and Anthropic Messages.
Model URL
Enter the model URL obtained from the model provider. Note that only the base URL of the API is required.
For example, if the model address is https://open.bigmodel.cn/api/paas/v4/chat/completions, enter https://open.bigmodel.cn/api/paas/v4 here.
If you enable Use Full URL, enter the full URL of the model.
Model ID
Enter the model ID obtained from the model provider. Its value contains a maximum of 64 characters. The model ID cannot be modified after the model is added. For example, glm-5.2.
Model Name
Enter the name of a model displayed in CodeArts Agent. The name contains a maximum of 64 characters.
API Key
Enter the API key obtained from the model provider.
Context Window Tokens
Optional. Set the input tokens and output tokens.
- Input: indicates the maximum number of tokens that can be received by the model in a single request.
- Output: indicates the maximum number of tokens that can be delivered by the model in a single response.
Description
Optional. Enter a maximum of 200 characters to describe the custom model.
- Click Add Model. The system automatically performs a model connectivity test. A real request will be initiated during the test, consuming a small number of model tokens.
- If the test fails, the page displays possible causes. Modify the model parameters based on the information provided, and then click Test and Add to run the connectivity test again. If you are sure that the model parameters are correct, you can also click Save Directly to finish the model addition.
- If the test passes, the new model is displayed in the Custom model list and is enabled by default.
Using Custom Models
In the lower part of the chat box, choose Built-in > Agent, or choose Custom and select a custom model from the model list on the right.
Managing Custom Models
After creating custom models, enterprise members can edit, delete, and disable them as required.
| Operation | Procedure |
|---|---|
| Editing a custom model |
|
| Deleting a custom model |
|
| Enabling/Disabling a custom model | In the Custom area, click Once disabled, custom models will no longer show up in the model list in the lower part of the chat box. |
What is your overall rating for this page?
Thank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot



