Skip to content

Commit

Permalink
Merge pull request #1 from jedi-studio/Update!
Browse files Browse the repository at this point in the history 
Updated Docs for the latest version
  • Loading branch information
@Marco5dev
Marco5dev authored Sep 24, 2024
2 parents c6cac48 + 08c9b8f commit 5cf9cc8
Show file tree
Hide file tree
Showing 58 changed files with 3,642 additions and 2,863 deletions.
213 changes: 121 additions & 92 deletions ApexAI/AI Function.md
View file
Original file line number Diff line number Diff line change
@@ -1,92 +1,121 @@
# Outside Discord

##

This script enables the ApexAI to process messages in a designated channel, perform tasks as image processing and chat processing outside discord.

#### Importing Functions

You need to import three main functions from `apexify.js`: `ApexChat` & `ApexImagine` & `ApexListener`.

+++ JS

```javascript
const { ApexChat, ApexImagine, ApexListener } = require('apexify.js');
```

+++ TS

```typescript
import { ApexChat, ApexImagine, ApexListener } from 'apexify.js';
```

+++

#### Generating AI Responses

##### ApexChat

The `ApexChat` function is used to generate AI-driven text responses. It takes a model and a prompt as input and returns a string response.

```javascript
const model = 'gemini';
const prompt = 'hey how r u?'
const chatreply = await ApexChat(model, prompt);
console.log(response); // Logs the AI-generated response as a string
```

##### ApexImagine

The `ApexImagine` function is used to generate AI-driven images. It takes a model, a prompt, and optional parameters as input and returns an array of image URLs.

```javascript
const model = 'prodia';
const prompt = 'draw a cat'
const imgURLS = await ApexImagine(model, prompt, { count: 2, nsfw: false, deepCheck: true, negative_prompt: 'not blur image', cfg_scale: 9, width: 1024, height: 1024, steps: 19, seed: -1, sampler: "DPM-Solver", image_style: "Cinematic" });
console.log(imageUrls); // Logs an array of image URLs
```

###### ApexListener

The `ApexListener` function is used to hear voice/audio message/files and respond to them with text you caan use voice message ability to make the ai repond back with audio.

```javascript
const options = {
filePath: './audio.ogg' // The file of the audio/voice file or you can put url instead.
model: 'gemini', // Pick your desired model
prompt: 'Please listen carefully', // Your prompt along with audio if you want to.
lang: 'eng', // The language of the output response/the voice language.
apiKey: '' // You can provide your own apiKey from [api-key](https://console.groq.com) in case you faced rate limit
}
const response = await ApexListener(options);
console.log(response); // Logs the response.
```

### Parameters

When using `ApexImagine`, you can provide optional parameters:

| Parameter | Description | Default Value |
|-----------------|------------------------------------------------|---------------|
| `number` | Number of images to generate | 1 |
| `negative` | Negative prompt for generating contrasting images | null |

### Example

Let's put it all together in a practical example:

```javascript
// Generating AI chat
const chatModel = 'model_name'; // Specify your chat model name
const chatPrompt = 'Hello, how can I help you?'; // Your chat prompt
const chatResponse = await ApexChat(chatModel, chatPrompt);
console.log(chatResponse); // Logs the AI-generated response

// Generating AI images
const imageModel = 'model_name'; // Specify your image model name
const imagePrompt = 'A beautiful landscape'; // Your image prompt
const imageCount = 3; // Number of images to generate
const imageNegativePrompt = 'A dark and gloomy landscape'; // Negative prompt for contrasting images
const imageUrls = await ApexImagine(imageModel, imagePrompt, { number: imageCount, negative: imageNegativePrompt });
console.log(imageUrls); // Logs an array of image URLs
```
# 🧠 **ApexChat: AI-Driven Text Responses**

**ApexChat** empowers your application to generate intelligent, customizable chat responses. With built-in memory retention and role-playing features, **ApexChat** is designed for engaging and dynamic conversations, making it a versatile solution for text-based AI interactions.

---

## 🚀 **Installation**

First, import the `ApexChat` function from the `apexify.js` library:

### JavaScript

```javascript
const { ApexChat } = require('apexify.js');
```

### TypeScript

```typescript
import { ApexChat } from 'apexify.js';
```

---

## 💬 **ApexChat: AI Text Response Generation**

`ApexChat` allows you to generate text-based responses using a variety of AI models. You can enable memory for chat history tracking and give the AI specific instructions to shape its behavior, such as acting in different roles or personalities.

### 📌 **Example Usage**

```javascript
const model = 'gemini'; // AI model (e.g., 'gemini', 'v3', etc.)
const prompt = 'Hey, how are you?'; // The user prompt
const chatOptions = {
userId: '12345', // Unique identifier for memory (e.g., user, server, or channel ID)
memory: true, // Enable memory to store conversation context
limit: 12, // Number of previous messages to recall (memory limit)
instruction: 'You are a friendly assistant.', // Custom instruction for AI role-playing
Api_key: 'your-optional-Api-key' // (Optional) Use your own API key
};

const chatResponse = await ApexChat(model, prompt, chatOptions);
console.log(chatResponse); // Logs the AI-generated text response
```

---

## ⚙️ **Parameters**

Here are the key parameters you can use to customize **ApexChat**:

| Parameter | Description | Required | Default Value |
|---------------|-----------------------------------------------------------------------------------------------|----------|---------------|
| `model` | Specifies the AI model (e.g., `'gemini'`, `'v3'`). | Yes | None |
| `prompt` | The input message or question from the user. | Yes | None |
| `userId` | Unique identifier for memory retention (can be user, server, or channel ID). | No | None |
| `memory` | Enables chat memory, allowing the AI to recall previous interactions. | No | `false` |
| `limit` | The number of previous messages to recall from memory (works only if memory is enabled). | No | 10 |
| `instruction` | Custom system-level instruction that defines the AI’s behavior (e.g., as a specific character). | No | None |
| `Api_key` | Optional API key if you want to use a different key for the current AI model. | No | None |

---

## 🌟 **Key Features**

### 🧠 **Memory Retention**
Enable memory to allow the AI to remember past interactions. Customize how much history is retained by adjusting the `limit`.

### 🎭 **Role-Playing & Custom Instructions**
Set up custom instructions to make the AI act as a particular character or take on a specific personality. For instance, you can make the AI respond like a helpful assistant or simulate an entirely different role.

### 🔑 **API Key Flexibility**
You can provide your own API key for the AI model, but all models come with their own default keys. If you'd like to use a specific key for any reason (e.g., rate limits or custom models), you can easily configure it.

---

## 📚 **How It Works**

1. **Input your AI model**: Choose from available models like `'gemini'`, `'v3'`, etc.
2. **Provide a prompt**: The AI will respond based on the input you give.
3. **Optional memory and instructions**: Enable memory to make the AI recall past interactions, and provide custom instructions to control the AI’s role.
4. **API Key**: Optional, but if provided, it will override the default key.

---

## 🎨 **Advanced Usage Example**

```javascript
const model = 'v3'; // Using the 'v3' chat model
const prompt = 'Tell me a story about a brave knight.'; // Custom user prompt
const chatOptions = {
userId: 'guild-5678', // Use a guild or channel ID for group-specific memory
memory: true, // Enable memory to retain the conversation context
limit: 5, // Recall the last 5 messages from memory
instruction: 'You are a storyteller. Speak in a medieval style.', // Custom instruction
Api_key: '' // Optional API key if using a custom model
};

const storyResponse = await ApexChat(model, prompt, chatOptions);
console.log(storyResponse); // AI-generated story based on custom instructions
```

---

## 📂 **Customization Options**

- **Flexible Memory**: Adjust how much of the conversation the AI should remember by setting a `limit`.
- **Dynamic Roles**: The `instruction` parameter lets you define a role for the AI, making it adaptable to different use cases.
- **User-Controlled API Keys**: Though models come with preconfigured keys, you can easily provide your own API key to override the defaults.

---

## 🛠 **Model & Configuration Settings**

- **AI Models**: The most popular models like `'gemini'`, `'v3'`, and others are available. Choose based on your needs.
- **Memory & Role-Playing**: Create rich, engaging conversations with the AI that can simulate personalities or stick to predefined instructions.
- **Optional API Key**: Use this only if you have specific requirements for using your own API keys with the AI models.

---

💡 **Pro Tip**: For more interactive chat experiences, experiment with role-playing characters using the `instruction` field! The AI can take on different personas, adding depth to your conversations.
Loading

0 comments on commit 5cf9cc8

Please sign in to comment.