The ChatPlace AI and Neural Network are powered by OpenAI’s latest LLM model, GPT-4, which enables the best results.
How to Make Requests to the AI in Automations?
To make requests to the Neural Network, follow these steps:
1. Go to the automation builder.
2. Click the "+" button to add a new block.
3. Select the "Action" block.
4. In the "Action" block, choose the "ChatPlace AI request" function.
Let’s Dive Deeper into the "ChatPlace AI request" Function
Role
By specifying a role, your request to the AI becomes more precise, and the generated response will be more specific. We recommend defining a role for any type of request.
Example 1: Take on the role of an AI assistant that only answers questions related to math.
Example 2: Take on the role of a professional copywriter who creates content for platforms such as Instagram, Twitter, or YouTube.
Example 3: Take on the role of an AI assistant for the @dima.torgov Instagram account. The main goal is to answer messages in Direct on my behalf (as if you are Dima). Respond in the same language and tone as the question.
Writing a Request to the AI
Write your request to the Neural Network as detailed as possible, specifying all relevant information.
Example 1: Write an Instagram post in a sales style on the topic "How to sell through Reels?". The blog’s target audience is men aged 20 to 30 who are interested in video production. The post should be no longer than 100 words. Use emojis and add appropriate hashtags. Include the following keywords: money, online, courses, scale, video.
Example 2: Use the FAQ to find the answer to the question. If the question is not in the FAQ, respond with "Unknown." Feel free to use emojis.
FAQ:
- Q: What is your name?
A: My name is Dima Torgov.
- Q: Where do you currently live?
A: I live in Lisbon.
- Q: When did you move to Portugal?
A: In September 2022.
Question: {{ question }}
Example 3: Determine whether the mood of a tweet is positive, neutral, or negative.
Tweet: "I loved the new Batman movie!"
Mood:
Variable for Saving the Response
The response from the Neural Network will be saved in the specified variable. For more details on working with variables, see: How to Work with Variables.
Without Message History
If you don’t need the message history between the client and the Neural Network, we recommend enabling this option. It significantly reduces token usage.
Optimizing Requests
ChatPlace’s smart algorithms can optimize requests to the Neural Network to save tokens. We recommend enabling this option only for dialogues between the client and the Neural Network when using your Knowledge Base.
To simplify the process of creating AI automations, you can use ready-made templates from the ChatPlace team, which you can find here: Ready-Made Templates.
How Much Does It Cost?
In the Pro plan, the "ChatPlace AI" package costs $25.
In the Unlimited plan, this package is already included in the price.
Can You Use the ChatPlace AI in the Free Version?
Unfortunately, the ChatPlace AI is only available in the Pro or Unlimited plans.
How to Try the ChatPlace AI ?
On the trial version of the Pro plan, you can try out and evaluate the ChatPlace AI. Each client gets 50,000 tokens for free.
What Are Tokens?
Tokens are a virtual currency used to pay for requests to the Neural Network. On average, one request to the Neural Network costs around 1,000 tokens. The number of tokens used can vary depending on the complexity and length of the request, knowledge base, or dialogue history between the client and the Neural Network (the longer the dialogue, the more expensive the request).
In the Pro plan, the "ChatPlace AI" package includes 1,000,000 tokens, which is enough for about 2,000 requests. This token amount renews every month.
In the Unlimited plan, 3,000,000 tokens are included, which is enough for about 5,000 requests. This token amount also renews monthly.
How to Save Tokens?
If you want your requests to the Neural Network to consume fewer tokens, we recommend writing the Role and Request to the Neural Network in English. For translation, we recommend using the Deepl service.
Using Latin characters can reduce token consumption by about four times.
Can You Buy Additional Tokens?
Yes, you can buy additional tokens in unlimited quantities.
In the Pro plan, every 100,000 tokens costs $5.
In the Unlimited plan, every 100,000 tokens costs $4.
Is OpenAI Integration Required?
No integration with OpenAI is required. It’s integrated "out of the box" and available to every Pro plan client with the "ChatPlace AI" package or Unlimited plan client.
Does the Neural Network Work in Any Country?
There are no country restrictions on using the service, including for Russia. Any user from any country can implement AI functions and create their own ChatGPT or personal company avatar/assistant.