Artifacts is a feature that can generates websites, scalable vector graphics, code snippets, etc. from your conversations and displays the generated content independently. You can preview the bots in a separate window in real time while chatting with ChatGOT, and flexibly edit and adjust them by natural language, which can greatly enhance your creation and interaction experience.
ChatGOT Bots supports for creating AI chatting bots based on different AI models by inputting prompts and setting skills. You can use ChatGOT Bots to fulfill various needs, such as writing, data analysis, translation, and more, all will be easily realized.
● Diverse Collection of Bots
ChatGOT offers a diverse range of Bots, including Psychologist, Physics Expert, Movie Critic, and more. You can directly choose to use these existing Bots without creating a new one.
● Diverse Selection of AI Models
When creating a Bot, ChatGOT provides a rich selection of AI models that you can flexibly configure to suit different needs and scenarios.
● Flexible setup of Bots skills
You can set various skills for bots in the description so that they can solve your problems more precisely. For example, you can set the following skills to fulfill writing needs: material gathering, language style interpretation, and structural profiling.
● Reuse & Secondary Editing
Bots you've created can be viewed in “My Creation,” supporting reuse for similar requirements. And they also can be edited for further refinement according to your needs at any time.
Introduction to Preset Bots
Click “Bots” to access the function page, and click “Explore” at the top of the chat list to navigate to the ChatGOT Bots Square page. Here you can browse all the Bots ChatGOT has preset for you, covering areas such as lifestyle, skills training, academic guidance, and more. Experience Bots like a psychological counselor, a travel planner, a speaking coach, and other characters.
Experience Preset Bots
In Bots Square, you can quickly find the Bot you want to experience by using the search function. When you tap on a Bot, you can view the following information:
After reviewing all the above information, click on “Chat Now” to quickly start a conversation.
Steps of Creation
1. Start creating: Click "Create Bot" to start creating your first Bot.
2. Set a name
3. Select an AI Model
4. Set a prompt
Setting a prompt is a important step that determines the Bot’s identity and how it responds to you. Prompts are instructions that guide the Large Language Model (LLM) to generate specific outputs. For example, if you need the Bot to write a piece of prose, you can set the following prompt:
Write a thoughtful and reflective essay on the importance of personal growth, following a clear structure:
Use a mix of personal experiences, philosophical insights, and relevant anecdotes to enrich the essay.
5. Add Key Skills
After the Bot’s persona is set, the next step is to add the skills required to accomplish the task based on your needs. Continuing with the previous example, you can assign it the following abilities:
You are qualified the following skills:
After completing all the settings, click "Create Bot" to finalize the creation of your Bot and publish it.
6. Test Your Bot
Once the Bot is created, you will automatically be redirected to the “My Creation” page, where your Bot creation history is displayed by time. Click on the newly created Bot to start a conversation and test it. ChatGOT also supports editing and deleting if you are not satisfied with its answers.
Overview of Functions
You can create and chat with bots from different AI models based on your needs. Bots can provide you with more accurate answers by selecting an AI model, setting prompts, adding skill descriptions, and more.
Selecting an AI Model
ChatGOT provides a variety of AI models to choose from, allowing you to select a model for the Bots to fit your needs.
➢ GPT Series
Pros:
GPT models, particularly GPT-4 and GPT-3.5, support multiple rounds of dialogue and complex reasoning tasks. With strong language logic and fluency, they can better understand the context of conversations with users.
Recommended Use Cases:
Content creation, educational purposes, and more.
➢ Claude Series (Anthropic)
Pros:
The Claude models focus more on the safety of the conversation and emphasizes "positive" conversation and "correct" output. They are better at controlling content during multiple rounds of conversation to avoid deviating from the topic.
Recommended Use Cases:
Mental health support, sentiment analysis and emotional processing.
➢ Llama Series (Meta)
Pros:
The Llama series (e.g., Llama 2) is optimized for the open source community and enterprise needs, focusing on efficient computation with fast response times, as well as powerful reasoning capabilities for large-scale data processing.
Recommended Use Cases:
Data analysis, business report writing, etc.
➢ Gemini Series (Google DeepMind)
Pros:
The Gemini models combine Google’s search engine technology and large-scale language models, specializing in information retrieval, and are able to quickly extract and process large-scale information.
Recommended Use Cases:
Academic research, legal, medical, financial and other specialized areas.
➢ Mistral Series (Mistral AI)
Pros:
The Mistral models focus on flexibility, feature multilingual support, and are capable of handling the input and output of information in multiple languages.
Recommended Use Cases:
Localization, translation, and other cross-language needs.
➢PaLM 2 series (Google)
Pros:
The PaLM 2 models possess multi-task learning capabilities and can handle various tasks across multiple fields simultaneously. They excel in areas such as speech recognition, natural language processing, and other domains.
Recommended Use Cases:
Speech recognition, cross-domain customer service, consulting, etc.
➢Hermes Series (Meta)
Pros:
The Hermes models focus on efficiency and performance optimization, excelling in low-latency, high-throughput applications. They also have outstanding capabilities in interactivity and creative content generation.
Applications:
Suitable for conversational applications that require fast response and high interactivity.
Recommended Use Cases:
Game interaction, virtual assistant, etc.
➢Qwen series (Tencent)
Pros:
Qwen models specialize in localization optimization for Chinese and Asia-Pacific markets, especially emphasis on Chinese language processing and cultural background. They can handle more delicate emotions and cultural differences in conversational interactions.
Recommended Use Cases:
They are recommended for dialogue systems in the Chinese market.
...
Prompts
Prompts define the persona and response logic of Bots, such as making it a psychological mentor, English teacher, thesis advisor, etc. Prompts determine how Bots interact with and respond to user queries.
Adding Skills Description
By adding key skill descriptions, you can set the Bot more accurately, enabling it to better understand your needs and provide more accurate responses. For example, if you want your Bot to produce more localized translations, you can specify the following skills in the description:
You are required to:
● Automatically detect and identify the input language
● Research and incorporate relevant cultural context from the identified region
● Process and adapt regional-specific terminology
● Automatically adjust the tone of voice
...
Refining
After testing the configured Bot, ChatGOT offers an advanced editing feature for further optimization and refinement.
Prompts are natural language instructions that define an AI’s behavior and response patterns, essentially establishing its personality and interaction logic. Setting effective prompts is a crucial step in bot creation. Well-designed prompts enable the Bot to better understand user requirements and execute tasks more efficiently and accurately.
Key Points
● Clear Task Instructions: Prompts should clearly define the task for the AI to execute.
Example: “Please translate the following text into [target language]”
● Personality Configuration : Define the Bot's persona and communication style to ensure responses align with desired user interaction patterns.
Example: “You are an expert translator specializing in professional documentation. Please translate the following text into [target language], maintaining a formal tone and technical accuracy.”
● Detailed Requirements Specification : Provide specific requirements or details to ensure the Bot generates targeted responses.
Example: "Please analyze the provided text and:
○ Extract the five key points
○ Present each point in concise bullet points
○ Provide a brief explanation (2-3 sentences) for each point
○ Maintain clarity and logical flow throughout"
● Output Format Specification: If you need the Bot to output content in a specific format, such as lists, paragraphs, tables, etc., define precise formatting requirements to ensure the Bot deliverables match desired presentation structures.
Example: "Please format the translation as follows:
○ Title: [Centered, Bold]
○ Executive Summary: [Single paragraph, max 100 words]
○ Main Content:
■ Section headers [Bold, left-aligned]
■ Subsections [Indented, bullet points]
○ Conclusion: [Italicized, separate paragraph]"
● Dialogue Flow Configuration: Establish coherent conversation patterns and contextual awareness to maintain logical discussion progression.
Example: "Building on our previous discussion:
○ Expand on [specific point] mentioned above
○ Analyze the relationship between [Topic A] and [Topic B]
○ Provide additional insights related to [previous context]
○ Connect this information with [earlier discussion point]"
Other Writing Guidelines
● Goal Setting and Expectations: Clearly defining the task’s objectives helps the Bot better understand and meet your expectations.
Example: “I’m developing a marketing plan for cross-border e-commerce with the goal of increasing conversion rates. Please provide specific recommendations based on this objective.”
● Role Definition: Clearly defining the Bot’s role, responsibilities, and communication style enables more precise responses aligned with your instructions.
Example: “You are a writing expert specializing in philosophical essays based on personal development experiences.”
● Operating Logic: Specify the Bot’s operational logic and response protocols for different use cases.
Example: “When users inquire about essential elements of effective writing, you may suggest using Google Search to gather relevant information.”
● Response Parameters: Setting clear boundaries for the Bot’s responses ensures answers remain focused and relevant to your inquiry.
Example: “Please limit your response to discussing the essential elements of contemporary prose.”
