8 Nov 202320:52

TLDRThe video introduces a solution for those seeking uncensored AI interaction, presenting the Uber Bata Text Generation Web UI. This software allows users to run AI text models locally on their computers, enabling private and unfiltered conversations. Installation is straightforward, with a one-click installer for Patreon supporters or a manual process involving Git for Windows. The interface enables users to download AI models from Hugging Face, with options for CPU and GPU-based models. The video also demonstrates how to use the software for roleplay with virtual characters, text-to-speech, and even image analysis. The creator offers exclusive characters and technical support for Patreon supporters.


  • ๐Ÿš€ Introduction of a new solution for running uncensored AI models locally on one's computer.
  • ๐Ÿ› ๏ธ Utilization of a software called 'uTab Text Generation Web UI' for interfacing with AI text models.
  • ๐Ÿ’ป Installation methods provided, including a one-click installer for Patreon supporters and a manual installation process.
  • ๐Ÿ”„ Use of 'Git for Windows' to clone the repository and install the necessary files.
  • ๐Ÿ“ Creation of a new folder to house the software and avoid issues with spaces in the folder name.
  • ๐Ÿ–ฑ๏ธ Instructions on running the 'start_windows.bat' file to initiate the setup and download additional files.
  • ๐Ÿ”ข Selection of GPU options during installation, with guidance on choosing the appropriate model based on available hardware.
  • ๐ŸŽฏ Guidance on downloading AI models from 'Hugging Face Doco' and understanding the different model formats (e.g., gguf, awq, gptq).
  • ๐Ÿค– Explanation of model parameters and their impact on performance and resource usage.
  • ๐ŸŽญ Roleplay capabilities with virtual characters, offering a variety of personalities and interactive experiences.
  • ๐Ÿ”Š Integration of text-to-speech and speech-to-text extensions for a more immersive experience with AI characters.

Q & A

  • What is the main purpose of the software Uba Tex mentioned in the transcript?

    -The main purpose of Uba Tex is to provide a platform for users to run uncensored large language models on their local computers using an interface called Text Generation Web UI.

  • How can users install Uba Tex on their computers?

    -Users can install Uba Tex either through a one-click installer provided for Patreon supporters or by manually installing it using G for Windows to clone the repository and follow the steps outlined in the transcript.

  • What is the role of the 'start windows.bat' and 'update windows.bat' files in the Uba Tex installation process?

    -The 'start windows.bat' file is used to run the installation and launch the web UI, while the 'update windows.bat' file is used to update the software.

  • What is Hugging Face Doco and how is it used in the process?

    -Hugging Face Doco is a website where users can find a variety of AI models, datasets, and test AI software for free. It is used in the process to download AI models for use with the Uba Tex software.

  • What are the two file formats for AI models mentioned in the transcript, and what determines the choice between them?

    -The two file formats mentioned are 'awq' and 'gptq'. The choice between them depends on whether the user has a GPU. 'awq' is the preferred format for those with a GPU, while 'gptq' is an older format.

  • What are the recommended models for beginners with and without a GPU?

    -For beginners without a GPU, the recommended model is 'Mistral 7B', which requires around 7 GB of RAM. For those with a GPU, the 'lat 2 30 billion' model is recommended, using around 7 GB of VRAM.

  • How does the character gallery in the Uba Tex interface enhance the user experience?

    -The character gallery allows users to choose from a variety of virtual characters, each with their own personality and way of speaking, to have more engaging and personalized conversations.

  • What are the two text-to-speech options provided in the transcript and how do they differ?

    -The two text-to-speech options are 11 Labs and Srow. 11 Labs offers high-quality voices but has a limited free monthly quota, while Srow provides free voices that, while not as high quality, allow for unlimited use.

  • How can users have a conversation with a virtual character using their microphone and receive audio responses?

    -Users can enable the whisper STT extension and apply the flags to restart the web UI. This allows them to use their microphone to communicate with the virtual character, and the character's responses will be played as audio.

  • What special model is needed to analyze and describe images, and how is it used in the Uba Tex interface?

    -The Lava model is needed to analyze and describe images. Users can select the instruct mode and upload an image to have the model generate a description and engage in a conversation about the image.

  • What is the one-click installer mentioned in the transcript and where can users access it?

    -The one-click installer is a simplified installation method for Uba Tex that automatically downloads and installs everything needed. Users can access it by supporting the creator on Patreon.



๐Ÿš€ Introduction to Uncensored AI and UbaTec

The paragraph introduces the concept of uncensored AI and presents a solution for users seeking privacy in their conversations. It highlights the limitations of existing AI models like ChatGPT, which are controlled by large corporations and subject to censorship. The speaker, KA Overlo, introduces a tool called UbaTec, which allows users to run uncensored large language models on their local computers using their own microphones for interactive conversations without corporate surveillance.


๐Ÿ› ๏ธ Installation of UbaTec and Text Generation Web UI

This paragraph provides a step-by-step guide on how to install UbaTec and the Text Generation Web UI. It outlines two methods: a simple one-click installer for Patreon supporters and a manual installation process for others. The manual installation involves downloading G for Windows, cloning a repository, and following a series of commands to set up the software. The paragraph also explains how to run the installation and update batch files, and what to do if a GPU is not present.


๐Ÿ” Selecting and Downloading AI Models

The speaker discusses the process of selecting and downloading AI models, emphasizing the importance of choosing the right model based on available hardware resources like GPU or CPU. It explains the difference between various model formats (GG UF, awq, and gptq) and recommends appropriate models for different hardware configurations. The paragraph also introduces Hugging Face Doco as a source for AI models and provides guidance on how to download and use these models within the Text Generation Web UI.


๐ŸŽญ Interactive Conversations with Virtual Characters

This section delves into the interactive capabilities of the Text Generation Web UI, allowing users to engage in conversations with virtual characters. It describes how to use the software to roleplay as different characters and emphasizes the freedom from restrictions that come with using uncensored AI. The paragraph also touches on the use of text-to-speech and speech-to-text extensions to facilitate real-time, interactive dialogues with AI characters.


๐ŸŒ Image Analysis and Multimodal Extensions

The final paragraph discusses the advanced feature of image analysis using the Text Generation Web UI. It explains how to download and use specific models like Lava for image recognition and analysis. The speaker guides users on how to set up multimodal extensions and use the instruct mode for image-based conversations. The paragraph concludes by encouraging users to explore the full potential of the web UI and offers exclusive characters for Patreon supporters.

๐ŸŽ‰ Conclusion and Final Thoughts

In the concluding paragraph, the speaker wraps up the tutorial on installing and using the UbaTec Text Generation Web UI. They offer a one-click installer and technical support for Patreon supporters and encourage viewers to try the software for themselves. The speaker also thanks the audience for watching, asks for likes and subscriptions, and acknowledges the support of Patreon supporters.



๐Ÿ’กuncensored roleplay

The term 'uncensored roleplay' refers to the creation and interaction with virtual characters or scenarios without restrictions or limitations on content, which is a central theme of the video. The video discusses how the software being introduced allows for such roleplay by running on a local computer, thus bypassing the restrictions that might be imposed by large corporations or AI models with censorship.

๐Ÿ’กAI models

AI models are the artificial intelligence systems used for various tasks, such as text generation, voice recognition, and more. In the context of the video, AI models are the core components that enable the text-to-speech, character interaction, and image analysis features. The video explains how users can download and utilize different AI models for these purposes.

๐Ÿ’กlocal computer

Refers to a personal computer that is not part of a larger network or central system. In the video, running AI models on a local computer is emphasized as a way to maintain privacy and avoid potential surveillance or censorship by big corporations. It is the platform where users can install and use the software for uncensored roleplay and AI interactions.

๐Ÿ’กHugging Face

Hugging Face is an open-source platform that provides a wide range of AI models, datasets, and tools for developers and researchers. In the video, it is mentioned as a source for downloading AI models, which is essential for users to access and utilize various AI capabilities within the software being discussed.


GPU stands for Graphics Processing Unit, a specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. In the context of the video, having a GPU can significantly enhance the performance of running AI models, as it allows for more complex and resource-intensive models to be used.


CPU stands for Central Processing Unit, which is the primary component of a computer that performs most of the processing inside the computer. It is responsible for executing instructions of a computer program, and in the context of the video, it is mentioned as an alternative to a GPU for running AI models, albeit with limitations in terms of processing power and efficiency.


Text-to-speech is a technology that converts written text into spoken words using synthetic voices. In the video, this technology is used to give AI characters a voice, allowing users to not only type interactions but also hear responses audibly, enhancing the immersive experience.

๐Ÿ’กmultimodel capabilities

Multimodel capabilities refer to the software's ability to handle and integrate multiple AI models simultaneously for different tasks. In the context of the video, this allows users to combine various AI functionalities, such as text generation, image analysis, and voice interaction, into a cohesive experience.

๐Ÿ’กPatreon supporters

Patreon supporters are individuals who financially contribute to a content creator's work on the Patreon platform, often receiving exclusive content or features in return. In the video, the creator mentions offering a one-click installer and technical support to their Patreon supporters, which is an example of exclusive content provided to those who support the creator's work.

๐Ÿ’กweb UI

Web UI stands for web user interface, which is the platform or interface through which users interact with the software. In the video, the web UI is the central component of the software being discussed, allowing users to run AI models, interact with virtual characters, and utilize various features without the need for external dependencies.


Introducing a solution for running uncensored large language models on a local computer, free from big corporations' surveillance.

The use of uaba Tex, generation web UI, a software that allows running any AI text models without restrictions.

A one-click installer for Patreon supporters simplifies the process of downloading and installing the necessary software.

Manual installation instructions for G for Windows to clone the repository and set up the environment.

Detailed steps for cloning the repository and installing the software using command prompt.

The option to run AI models using a CPU or GPU, depending on the user's hardware capabilities.

Guidance on selecting the appropriate AI model based on the available GPU or CPU resources.

Explanation of different AI model formats like GG UF, awq, and gptq, and when to use each format.

Instructions on downloading AI models from huggingface doco, a platform with a variety of AI models and datasets.

Recommendations for specific AI models like Mistral 7B and lat 2 30 billion for different user needs.

A step-by-step guide on how to download and load AI models into the web UI.

The ability to converse with virtual characters, each with unique personalities and conversation styles.

Inclusion of text-to-speech and speech-to-text extensions for a more interactive and immersive experience.

The integration of 11 Labs and srow for high-quality text-to-speech and speech-to-text without incurring costs.

Exclusive access to a special character, Sandra the loving girlfriend, for Patreon supporters.

The capability to analyze and discuss images using the web UI, similar to CH GPT Vision.

Technical support and a one-click installer available for users facing issues, provided through the Patreon page.

The potential for endless entertainment and creative applications with the web UI, beyond traditional AI interactions.