Greetings, thank you for reaching out.

Currently, there is no direct link to obtain the extended license. It should be selected alongside the product by clicking the dropdown arrow to switch from the Regular License to the Extended License. Please note that the extended license is not eligible for promotional offers like the regular license.

also from your experience, is the generate images spend much money? can I disable the images, only to use it for texts only?

Hello, thank you for your contact.

To adjust character limits, you should do this directly in the “character.json” file located in the “json” folder.

Use the “chat_minlength” field to set the minimum number of characters and “chat_maxlength” to define the maximum number of characters.

Regarding text typing, we have an FAQ at the bottom of the page that discusses this in the section “why isn’t my text streaming function working?”

Text streaming refers to the capability of obtaining text responses in segments or parts. With this functionality, the Chat GPT API divides and delivers the response in smaller portions, allowing the user to view the answers without having to wait for the complete conclusion of the request.

You can find more on the subject here, along with some tips on how to enable it on your server with your hosting support: https://documentation.polartemplates.com/aigency-saas/text-streaming

hello can you answer.

Hello! The most effective approach would be to adopt the “fine-tuning” model. However, as specified in the product description, this template does not use “fine-tuning.”

But you have the option to input details and training information about your product directly into the prompt in a Frequently Asked Questions (FAQ) format.

For example, if your product is a cleaning item, you can define the prompt as follows:

“You are a virtual assistant tasked with answering questions about my cleaning product. The product is called X, costs Y, and is used for specific purposes…” This way, by providing as many details as possible, the bot will use this information to respond to user queries.

Please remember that the ChatGPT 3.5 Turbo model has a token limit of 4000, allowing for approximately a prompt with up to 1500 characters. However, if you need more space, you can opt for the ChatGPT Turbo 16k model, which offers capacity for four times more tokens, allowing for more detailed inserts in the prompt.

If the bot doesn’t respond effectively, we recommend rewriting the prompt and testing it. You can test the prompt through the OpenAI playground.

https://platform.openai.com/playground

Hello, we currently do not have plans for React. Thank you for reaching out.

Hello! Thank you for getting in touch. Certainly, it’s possible to perform this action using the animal’s training prompt. When editing the animal’s prompt, simply include these instructions clearly. For instance: “When the user inquires about how to register (“how to register”), direct the system to display the personalized link.”

Hello friend, the latest update was the upgrade to models GPT-4 and GPT-3.5 Turbo with 16k tokens. Currently, we are working on updates for this model:https://codecanyon.net/item/chat-gpt3-openai-aigency-html-5/44616793

Hello, my friend, thank you for reaching out. As mentioned in the product description, we exclusively use the GPT 3.5 Turbo or text-davinci-003 models. Unfortunately, the Babbage model is not available. However, we are planning to upgrade our model to use the GPT-4 and also the new GPT 3.5 Turbo 16k.

Hello, it is possible to configure the minimum length of the question to the desired value through a configuration file. If you have any other questions, feel free to ask. Also, take advantage of our special discount starting from June 2nd when purchasing the template.

Hello friend, we have davinci and we have implemented chat model gpt 3.5 turbo, which is 10x cheaper than davinci, we recommend this model.

I also invite you to see our new update with SAAS model https://codecanyon.net/item/chat-gpt3-openai-aigency-html-5/44616793

Hi, so the gpt 3.5 turbo is available for this one and for aigency? A colleague has this one, not aigency, and it shows error when she tries to set it with gpt 3.5 turbo (in the api.php file), so that’s why I asked, because it hasn’t work with gpt 3.5 just davinci.

Yes, maybe it has the old version, when we released this model in the beginning it only supported davinci, however after many updates we put gpt 3.5 turbo for this model.

The aigency model is now available also the gpt4.

Hello friend, we launched the CMS on the 26th for this model.

https://codecanyon.net/item/chat-gpt3-openai-aigency-html-5/44616793

Hello, Yes, here is a list of available voices.

https://cloud.google.com/speech-to-text/docs/speech-to-text-supported-languages?hl=en-us

Hello friend, it is possible as long as you specify it in the training prompt. In the case of the demo example, no restrictions were imposed, but you could write during the training, for example, “Only answer questions about Medicine, do not respond to questions on any other topic.” Then the AI will follow that instruction.

This is awesome Thank you!

Hello friend, please test by this link: https://smartanimals.polartemplates.com/v1/

We will update the demo link, it’s an old version.

Perhaps such errors occur because you are running out of requests?

Thanks for the quick response. Everything works on this link, very fast. I checked the old link again, it doesn’t work. My country doesn’t matter, requests are sent by the server and not by the user.

I thank you. Are you using Safari? Maybe due to the preview link, the Envato iframe is blocking something, I’m not sure why it’s not working for you there via preview.

Anyway, I’m glad you were able to test it. Thank you.

I’m not viewing in an iframe, the browser is Chrome.

try { tokens = JSON.parse(data); } catch (err) { console.error(`Error parsing SSE data as JSON: ${err}`); return; }

data: {“error”: “[ERROR]”,message allowed to POST on /v1/engines/completions. (HINT: Perhaps you meant to use a different HTTP method?)”} data: {“error”: “[ERROR]”,message writing body (0 != 218)”}

Hello, please send us your website by email so we can analyze it. However, when dealing with issues related to this error, it is possible that your API key credits have expired, which prevents the reception of a valid JSON response. Anyway, please send us your website so we can investigate the issue. My email is support@polargames.com.

Hello, the latest version within the zip file is _v3.0 (last), and it is from this version that we will include the next updates.

The old versions are available only because some clients requested them, however, version 3.0 is the most up-to-date.

Hello, the response speed of the API depends on various factors such as network delay and potential communication issues with the API. Some locations in the world may receive a faster response, while others may experience delays or no response at all due to API restrictions imposed in certain countries.

To check the API status, it is recommended to visit the website https://status.openai.com/. For example, two days ago at a specific time, the API experienced some issues, which could have caused delays or other problems.

Conversation history can also affect response speed. However, in the model we use, the sending of conversation history to the API is limited to the last eight conversations, but this value can be changed.

Personally, I have conversations stored since last month and do not experience any delay issues, except on days when the API is undergoing maintenance.

Hey, don’t worry i found it…. but i can’t find the privacy policy. how can i edit the privacy policy?

Hello, to change the logo, you can do so in the “img” folder, where you will find a file called “logo.svg”. Regarding the text of the privacy policy, you need to edit the “index.html” file. You can open this file with your preferred code editor or simply use Notepad for this task.

Thank you very much sir,

Please, one last request, Can you help us by providing sample code to embed in the Api.php file to record each chat question and answer?

I mean save in a database for example with sql

Hello! Each developer can use a different library or framework to handle specific tasks. For example, some may choose to use Laravel, CodeIgniter or Symfony, each with its own method for performing CRUD (Create, Read, Update, Delete) operations on databases. However, the most common method, without the use of frameworks, is to use the PDO extension of PHP. Unfortunately, I cannot provide code here, but I suggest that you research the topic “CRUD in PHP using PDO”. This way, you will already be able to achieve what you want.

Hello, I tested it here and it seems normal. Could you try testing it again? link:https://smartanimals.polartemplates.com/v1/ Thank you.