This morning, Elon Musk announced that the API service for xAI has been launched.
Our engineering team immediately began supporting xAI in AIGNE. A unique design of AIGNE allows us to implement adapters for supporting any large language model or AI content generation model as an Agent. As a result, we only need to develop and debug directly within AIGNE, making it very simple and convenient!
In the future, any developer with some basic programming experience will be able to write adapters for any model and publish them as Blocklets in our Blocklet Store, enhancing AIGNE's model support capabilities.
How to use xAI in AIGNE?#
Install the xAI adapter Blocklet#
The adapter Blocklet we developed for xAI is already available in the Blocklet Store. However, you only need to install these adapters if you are running an independent instance of AIGNE. If you are using the online service of AIGNE, you can skip this step.
Apply xAI API Key#
Since the current xAI API service is still in the beta testing phase, there are significant limitations on the allowed access volume. Therefore, for users who need to use xAI, you need to apply for the xAI API service yourself and create an API Key. Please follow the xAI documentation below to create an account and create an API Key (you will need a credit card to pay for the xAI).
After xAI officially launches its services in the future, we will directly support xAI in AIGNE, similar to OpenAI and Google Gemini, allowing users to access all services through ABT. However, during the current testing phase, users who enjoy trying new services will need to apply for their own xAI API Key.
Config xAI in AIGNE Project#
In the AIGNE project, clicking on the model selection configuration button of any Agent using a large language model allows you to choose different model services (Provider) and select the xAI LLM Adapter.
You will see the list of LLM adapters below, among which is xAI. This is what the xAI adapter Blocklet mentioned above provides. In the online version of AIGNE, we have already installed this Blocklet, so you can select it directly. If you are running your own instance of AIGNE, you need to install the xAI adapter first to see this option.
After you select the xAI adapter, you need to provide the API Key that you just obtained from the xAI application.
Use xAI in Agent#
After the above configuration, you can use xAI in the Agent, and its usage is exactly the same as other models you are already familiar with! You can try different prompts to test the capabilities of the xAI grok model. In AIGNE, you can mix and match various different models within the same project and easily switch between different large language models. In the following simple example, you can see that xAI is already working properly and has returned its own basic information.
Through the above steps, you can now use the latest released xAI in AIGNE. We will follow up on the progress of xAI and continuously make improvements!