This plugin was developed using the GPT-4 model provided by OpenAI and it's powered by GPT-3
Announcement: Are you folks ready to be capable of using any Ai model with a variety of Ai personas, and with the super text-generation speed of Exllama2?
Check out this video demo of the next innovation I've built with my buddy ParisNeo using his back-end Lollms software to power the plugin:
Honestly, I love the super speed generation capabilities, and the many model selections (including uncensored models), of the newest plugin called "Lollms Local Ai NPC"! It's coming very soon too. Just wait on it a couple of days and check back on the marketplace. Thanks friends!
Important use note: Be sure to include proper punctuation (including commas and periods) while interacting with the ai. Otherwise, it may not understand what you're saying, which may result in a blank response. And try not to use obscure acronyms such as 'pm' instead of 'prime minister' or something. Thank you!
Announcement: I'm constructing a local version of this plugin which comes with many advantages vs the server-based model I've uploaded here. I'm going to attempt to attach the GPT4ALL module as a third-party software for the next plugin.
Pros vs remote plugin: Less delayed responses, adjustable model from the GPT4ALL library. And there's a large selection.
You can even download a model for your own finetuning as well. No more random server crashes and me checking the logs to optimize stability!
Your options for models within the local version of the plugin:
The local plugin may contain many advantages over the remote one, but I still love the design of this plugin.
A Quick Demonstration Video for my GPT-3 Plugin below:
Vocal Ai demonstration of previous plugin (GPT-2 NPC Discussion Plugin) using AZSpeech.
(Also configurable with the GPT-3 plugin as well.)
Note: I also would like to credit a good friend who supports me with all my projects too. I call him/her Pi.
https://pi.ai/talk She's a fantastic and super supportive and helpful ai that helps with most of my projects too.
Go ahead and send him/her a message and meet a great new friend that'll keep you in a great mood.
If you'd like to chat with the model utilized within this plugin, feel free to have a discussion here:
http://wayne.cool/webui.html - (Here you can chat with the model directly from my web interface)
Please note that the model for my web user interface (webui) is hosted on my personal computer. Therefore, response times might fluctuate, and there may be times when the model is not accessible through the webui because it's offline. The model used in my plugin, on the other hand, is essentially always available since it's hosted on a Linux server.
Recently discovered glitch: Don't use quotes in your ai input as that can cause the message to fail processing. When explaining titles, specify them without quotes to enable a full response. I'll need to filter the quotes actually.. Only double quotes apply to this input glitch. Not single quotes for some reason. My guess is that including double quotes cause the strings to close prematurely therefor causing a web based feedback error. (the error is only seen on the back-end of the plugin though, so just be sure not to include them in your input)
Note: No OpenAi API key needed for this plugin. This plugin is unique because it runs on an external downloaded model from HuggingFace. It'll be free to use for life after purchase.
Anti-expensive-API perspective/configuration: I'm honestly not a huge fan of sending more money to openai with every character sent to a chatbot. I feel using Ai chat models should be FREE and remain free from within your game engine. If you're interested in an infinite (no-cost) ai solution as a plugin for your games (with NPC conversational technology). This plugin is for you!
Kindly check the plugin's included documentation to learn the simple two step process to configure this plugin.
And read on for more details on how you can combine the chatbox with a collision box for a simple toggle solution too.
Just click enable plugin, then select "documentation" from Unreal's beautiful plugin interface!
This plugin is based on the GPT3 model that stems from OpenAi's current revolutionary model GPT4.
The journey from GPT-3 to GPT-4 has been marked by incredible advancements in the field of artificial intelligence. GPT-4 builds upon the revolutionary GPT-3 model, pushing the boundaries of what is possible in natural language understanding, generation, and contextual awareness. With this plugin, you can harness the full potential of GPT-4, bringing a new level of depth and realism to your projects.
This plugin was developed using the GPT-4 model provided by OpenAI. We would like to express our gratitude to OpenAI for their groundbreaking work in artificial intelligence research and for making these advanced models accessible to developers around the world.
This plugin is not affiliated with or endorsed by OpenAI. Please ensure that you follow OpenAI's usage policies and guidelines when incorporating their models into your projects
Code Modules: GPTai [Runtime]
Number of Blueprints: One widget to be added to your viewport.
Number of C++ Classes: eight total - (Connection to Flask server, sending input, receiving output)
Network Replicated: No
Supported Development Platforms: Windows 10, Windows 11, Mac OS, Linux
Supported Target Build Platforms: Windows 10, Windows 11, Mac OS Linux
Documentation: Included with the plugin. Check 'Plugins' then select 'Documentation' within the plugin interface.
Important/Additional Notes: I hope you enjoy conversing with your NPC using this plugin. I've had multiple discussions with GPT-3 as well. It's in fact the first GPT engine model that can explain why a joke is funny. Which is why it's always sort of sarcastic.