Recently, 302.AI brought a big update-Custom Model. The launch of the custom model is to create a more open, convenient and practical platform for users.

Specifically, the custom model function implements the integration of third-party model APIs into the 302.AI (such as OpenRouter and silicon-based flow), adding the following advantages to the third-party model:

1、Quickly generate chat Robot: Users can access third-party large model APIs according to their own needs, and are not limited to resources in the platform. The entire integration process is extremely convenient, and it only takes a few simple steps to quickly complete the configuration and deploy the chatbot for online use.

2、Solve regional restrictions: When the model API is accessed, you can select the transit region to solve the problem of regional restrictions of some model suppliers.

3、One-click function expansion: Access to the model API can unlock 302.AI self-developed "ability bonus", such as network search, long-term memory, etc., and realize function expansion with one click.

Custom model price: Whether it is a custom model to generate a chatbot or call a custom model API, it is only $1.5 per month at $0.05/day, which is much lower than the cost of maintaining a server.

How to generate a chat robot using 302.AI custom model

一、Setting up a custom model

Enter 302.AI and click: Custom Model

Click Add Model and fill in the relevant configuration information.

(1) Provide preset APIBase

We have preset the API Base for Open Router and Silicon flow. Users who use these two services can directly select them and the system will automatically complete the filling.

(2) Support transit area selection

We also specifically support the transit region selection function, which can solve the problem of some model providers being restricted by IP region. For example, the Hong Kong region is often restricted when accessing the model API, so we can switch to the US region.

(3) One-click configuration information check

Equipped with a [Check] button, you can check whether the information is filled in correctly with one click.

1、Take the model API accessed to Open Router as an example:

Example Model:meta-llama/llama-3.3-70b-instruct:free

Select OpenRouter for API Base, and fill in the API Key, model ID, etc. Configure accordingly according to the model characteristics. For example, if the model itself has the image analysis function, check "Model supports image analysis", otherwise do not check it. After all settings are completed, click [Confirm];

2、Take the model of access to siliconflow as an example

Example Model: Siliconflow Deployment：DeepSeek-R1-Distill-Qwen-7B

Select Siliconflow for API base, fill in information such as API Key and Model ID, select the transit area as needed, and click [Confirm];

二、Creating a chat robot

(1) The model that has been successfully connected will be displayed in the custom model category. Select the model and click [Confirm].

(2) Click the [Create Chat bot] button:

(3) After entering the chat bot, send a message to start a conversation;

三、The connected model can achieve functional expansion with one click

After accessing the model to generate a chatbot, you can also implement image analysis, online search, reasoning and other functional expansions with one click.

1、Image analysis capabilities:

Take the connected Llama-3.3-70b-instruct as an example. This was originally a large plain text model. Connecting it to 302.AI can unlock the "model fusion" capability and start image analysis:

(Note: The image analysis function does not require special settings, the chatbot will be enabled by default)

2、One-click superposition of online search + reasoning function:

In addition, you can also use functions in a superimposed manner, such as enabling online search and reasoning functions at the same time. Just light up these two function buttons at the bottom of the dialog box:

The effects are compared as follows:

Original answer: Information lag.

After superimposing the online search + reasoning function, the answer is: correct.

四、To manage custom models:

(1) Edit: After completing the custom model, if there is any information that needs to be modified, click the edit symbol on the right side of the model name.

(2) Copy: After clicking Edit, you can quickly copy the configured custom model.

After copying:

(3) Delete: Custom models that are not frequently used can be deleted.

How to use 302.AI's API to call a custom model

一、Create an API Key

First, we need to create an API key that supports transferring custom models.

Detailed steps: API → API Keys → Enter API name → Enable custom model transfer → [Add API KEY]

Copy the API Key you just generated for later use;

Check the name of the custom model after transfer for easy calling:

二、API call custom model

You can add functions to the custom model API through the API supermarket or call the custom model in the tool API

1、Add new capabilities to the model:

Find relevant documents: API → API Market → LLM→ 302.AI → View Documents

Take adding the online search capability of a custom model as an example:

After entering the relevant document, click the [try it] button:

Click "Headers" and paste the generated API Key:

(Note: The custom model transfer function must be enabled manually, so the API Key needs to be pasted manually)

According to the documentation, add "-web-search" to the suffix of the custom model (such as the example meta-llama/llama-3.3-70b-instruct:-web-search) and enter the prompt word and click [Send] to add the network search capability:

2、Tool API calls custom models

Related documents: API → API Market → Tool API → Select tool as needed → View document

Take the AI quiz machine tool API to call a custom model as an example:

After clicking [Send], you also need to set the API Key. The steps are roughly similar to the above, so I won’t go into details here:

After setting the API Key, fill in the name of the custom model to be called, enter the prompt word and click [Send].

Summarize:

302.AI is committed to creating a more open and inclusive AI comprehensive platform for all users. Through this platform, users can not only enjoy a more convenient and smooth user experience, but also unlock more powerful and rich functions to truly maximize benefits.

302.AI is committed to creating a more open and inclusive AI comprehensive platform for all users. Through this platform, users can not only enjoy a more convenient and smooth user experience, but also unlock more powerful and rich functions to truly maximize benefits.

External Resource