Groq Web ^0.4.0-beta.0 Deprecated

Raycast Unblock offers Groq Web AI support. You can use Groq's models for free via their Web API.

WARNING

Unfortunately, the cost of bypassing Cloudflare protection on the NodeJS platform is unacceptable in this project. Therefore, we are unable to provide support for Groq Web AI.

---

DANGER

We found Cloudflare protection, which caused our program to fail to interact with api.groq.com normally. You can go to wibus-wee/raycast-unblock#72 for more details.

We will try to solve this problem as soon as possible. You can't use this feature until we fix it.

WARNING

This is an inverted API. Please use with caution.

The following content is no longer maintained and is for reference only.

Usage

1. Open https://groq.com/ and log in. Open the Network tab in the console. Refresh the page.

2. Locate the https://web.stytch.com/sdk/v1/sessions/authenticate request in the Network tab. Look for the Authorization header in the request.

3. Copy the value of the Authorization header, which will be in the format Basic xxx.

4. In your configuration file, set refreshToken in [AI.Groq] to the copied value (a placeholder value is already provided in the example configuration).

Configuration

The configuration for this feature includes the following parameters:

• refresh_token: This parameter refers to the value of the Authorization header. It is used to authenticate and authorize the user's access to the system.
• temperature: This parameter determines the variability and creativity of the output generated by the system. A higher temperature value will result in more random and diverse responses, while a lower value will produce more focused and deterministic output.
• max_tokens: This parameter specifies the maximum number of tokens that the output should contain. Tokens can be thought of as individual units of text, such as words or characters. Setting a higher value for max_tokens will result in longer output, while a lower value will generate shorter responses.
• default: The default model to use. Optional ^v0.5.0-beta.2

Example

[AI.Groq]
refresh_token = '<your refresh token>'
# temperature = 0.5
# max_tokens = 100