Integração do ChatGPT para Facebook e Instagram

On the StartMsg platform you can use the ChatGPT integration for Facebook e Instagram, which gives your business extra ways to serve customers with automated chats. Here are the simple steps to register and get your OpenAI ChatGPT Secret Key:
1. Go to the OpenAI website

Visit the OpenAI website at https://platform.openai.com/signup.

2. Create an account

If you do not have an account, choose Sign Up and fill in the required details. If you already have an account, you can log in by choosing Iniciar sessão.

3. Confirm your email address

After registering, check your email. OpenAI will send a confirmation message. Click the link in the email to confirm your account.

4. Go to API settings

Once logged in to your OpenAI account, go to the API settings page, where you will find the API Keys section.

5. Create a new Secret Key

Escolher Create new secret key, and OpenAI will generate a new API key. Copy this key and keep it in a safe place, as you will need it to integrate ChatGPT with StartMsg.

⚠️ Important: Never share your API key with other people or publicly, as it gives access to your OpenAI account.

6. Log in to StartMsg and add the API key

Now log in to your StartMsg account and enter the API key in the integrations section.

OpenAI settings on the StartMsg platform

1. OpenAI Secret Key

• What is it? The OpenAI Secret Key is your unique API key that lets the StartMsg platform use OpenAI services such as ChatGPT.
• Where to get it? After registering on the OpenAI platform and logging in to your account, go to the API settings page. There you can generate your API key.
• Why does it matter? The secret key ensures that the Integração do ChatGPT para Facebook e Instagram can work on your StartMsg platform. The API key is also required to enable OpenAI services.

Tip: Keep your API key in a safe place and do not share it with others.

2. Instruction to AI

• What is it? In this field you can enter additional instructions for the AI so it better understands your specific needs. For example, you can write that ChatGPT replies should be short and clear, or written in a certain tone.
• How to use it? For example: “Answer questions simply and clearly, as if talking to a new user.” Such an instruction helps ChatGPT tailor its replies the way you want.

3. Models

• What is it? OpenAI offers several artificial intelligence models with different capabilities and accuracy. The most commonly used models are GPT-3 and GPT-4.
• How to choose?
• GPT-3.5: Fast and economical, suitable for most tasks.
• GPT-4: More advanced, gives more accurate answers and is better suited to complex queries.
• Recommendation: Choose the model that best fits your business needs and budget.

Read more about ChatGPT models here; we recommend using ChatGPT 4o

4. Maximum Tokens

• What is it? This is a setting that defines the length of the answer the AI can provide. The higher the token count, the longer the AI can generate its answer.
• How to use it?
• Smaller numbers (e.g. 50-150) suit short answers and are usually used for fast, simple replies.
• Larger numbers (e.g. 500 or more) allow longer answers, but can be slower and require more resources.
• Recommendation: Start with a medium token count (200-300) and adjust this setting as needed based on the quality of the answers.

What is one token?

In OpenAI systems, tokens are the smallest units of information that the model reads and analyzes. Usually one token equals 4-5 letters or characters, and a short word such as “hello” can be one token. The sentence “How are you?” would be about 4 tokens.

Approximate token values:

• 1 token ≈ 4 letters or characters.
• 750 tokens ≈ roughly 500 words.
• 1,000 tokens ≈ roughly 750 words.

The token count depends on the text structure and language, so it can vary slightly.