Skip to main content

5 posts tagged with "KuwaOS"

View All Tags

· 2 min read
Yung-Hsiang Hu

Hello friends,

The TAIDE model was released today, and we are happy to release a customized Kuwa system for Windows that has the built-in TAIDE LX 7B Chat 4bit model.

info

Download link for the kuwa-taide-v0.2.0 single executable file: https://dl.kuwaai.org/kuwa-taide/v0.2.0/
kuwa-taide-v0.2.0 documentation: https://dl.kuwaai.org/kuwa-taide/v0.2.0/kuwa-taide-0415.pdf

This customized system is a self-extracting single executable file, and the TAIDE model is built-in as the default local model option, and it can be run in both CPU and GPU environments, allowing everyone to quickly and easily experience the effects of the TAIDE model and perform GenAI-related applications.

In addition, this system is customized based on the previously released v0.2.0-beta, so you can also use the TAIDE model for RAG applications. However, it is important to note that RAG will generate longer input, so it is recommended to use the GPU version for inference.

The Kuwa system and TAIDE model are still under development and improvement, so it is inevitable that they will be unstable. The content generated by this system is for reference only and does not guarantee its correctness. Users still need to verify it; please do not publish inappropriate dialogue content to avoid unexpected problems.

TAIDE official website: https://taide.tw/
Kuwa official website: https://kuwaai.org/zh-Hant/

· 4 min read
Yung-Hsiang Hu
info

This version does not include the TAIDE model itself, and a version pre-loaded with the TAIDE model is expected to be released after the TAIDE model is publicly available.

Hello to our community friends,

After collecting everyone's feedback, we plan to roll out the long-awaited RAG feature in v0.2.0. The RAG part has been internally tested, so we are releasing v0.2.0-beta to invite everyone to test it out and see if it meets your expectations.
In addition, this update also provides a way to connect with TAIDE API and TAIDE models.
At the same time, we have also adjusted the system installation script and fixed some known bugs, making the entire system more stable, easier to extend, and easier to use.
If you have any suggestions or if you think there is room for improvement, please let us know!

The details of this update are as follows:

Windows Portable Edition

  1. Adjust the model hosting method:
    • Enable Gemini Pro and ChatGPT APIs by default
    • Use Gemini Pro by default to launch RAG applications for WebQA and Document QA
  2. Deprecate env.bat and use run.bat to launch the executor instead
  3. Enhance executor functionality:
    • Allow direct configuration of execution instructions, parameters, and other information
    • Adjust init.bat to be a simple tool to help create run.bat. Users can also directly write run.bat to launch the required model
  4. Fix the error of a non-existent PHP download link in v0.1.0 (archived due to version update)
  5. Integrate RAG into the simplified launch framework of the Windows version
  6. Specify file path improvements:
    • In the executors folder of the Windows version, files will be specified using relative paths by default
  7. Fix the executor error of the Custom category
  8. Permission adjustment:
    • Only groups with Manage Tab permission can be directly assigned permission to use the model when the model is added
  9. Fix the issue in the Windows version where Redis uses localhost as the IP, causing DNS queries to be delayed by 2 seconds each time

Docker Edition

  1. Integrate RAG (Document QA / Web QA / DB QA) into the executor's Docker image and compose the system
  2. Provide a compose example of Gemini Pro using a global API key
  3. Complete the missing words in the document

Executor

  1. Provide a TAIDE API executor, which can be directly connected to the TAIDE API of TECO
  2. Port RAG executor (Document QA / Web QA / DB QA / Search QA) to the new framework
  3. Let the RAG executor support automatic model search, i18n, and interrupt generation

Kernel

  1. Provide an API to list currently available executors

Multi-chat

  1. Adjust the timeout waiting time for the no-message state:
    • Extend from 1 minute to 2 minutes to accommodate the waiting time for the RAG processing speed gap
  2. Fix the bug of AdminSeeder:
    • Fix the issue of granting duplicate model usage permissions
  3. Add the function of sending Kuwa tokens from the website to the executor:
    • To fix the past issue of unowned API tokens for RAG
  4. Add a method to adjust the default model image path:
    • Can be configured via LLM_DEFAULT_IMG in the .env file
  5. Fix the bug that the API in v0.1.0 could not be used normally

Known Issues and Limitations

  1. At present, the Windows version of Document QA can process files in .doc and .docx formats. However, due to library dependency issues, it may not be able to read certain .pdf files. If such a need arises, please consider utilizing the Linux version of Kuwa for PDF processing.
  2. RAG applications tend to generate long input. If only using the CPU-based version of the on-premises model, timeout errors can occur more easily. In this case, we recommend either using the cloud-based model, or using the GPU-based version of the on-premises model, and then using the RAG application.

· 2 min read
Ching-Pao Lin

Hello developers and users,

After receiving feedback from many users since the initial release, we are pleased to announce the stable release of v0.1.0. In this version, we have made some adjustments to the installation process for the Windows version. We have also simultaneously released a Docker version, allowing users to quickly install and adjust the environment structure as needed. Additionally, we have fixed some minor bugs that were known in previous versions.

Here are the main updates in this release:

Windows Portable Version

  1. Adjusted the model setup process to allow for easier configuration of multiple models.
  2. Fixed various errors that occurred when using MySQL or PostgreSQL.
  3. Readme updated for better completeness.

Docker Version

  1. Docker Compose can now be used to start the entire system and multiple Executors with a single command.
  2. Stable software stack selected, suitable for direct use in production environments.
  3. Modular design allows for the selection of Executor types and quantities to be launched freely.

Executor

  1. Added a command-line interface launcher that can start multiple Executors with one click, allowing common parameters such as Prompt template, System Prompt, and Generation config to be passed in as commands.
  2. Supports common on-premises model inference frameworks such as Huggingface Transformers and Llama.cpp.
  3. Supports inference services compatible with OpenAI API or Gemini-Pro API, such as vLLM, LiteLLM, etc.
  4. Packaged common functions into the Executor framework, such as automatic registration retry, automatic logout, automatic history record pruning, interrupt generation, etc.
  5. Packaged the Executor framework into a package for easy extension of Executors.
  6. Fixed a bug in the generation error of the llama.cpp executor.
  7. Changed the underlying framework to FastAPI to improve efficiency and stability.

Multi-chat

  1. Fixed bug causing website to jump to /stream route.
  2. Added default images for models.
  3. Fixed some minor bugs.
  4. Added more command-line tools for configuring the website.

For migration from older versions to the new version, please refer to this migration guide.

· 2 min read
Ching-Pao Lin

This is a tutorial for updating from the initial version to the stable version v0.1.0.

  1. First, clone the repository using git clone https://github.com/kuwaai/genai-os.git --tag v0.1.0, or download and extract it from here to get a clean copy of the v0.1.0 project.
  2. Here, the old version of the project is referred to as the old folder, and the newly obtained version is referred to as the new folder. If you have these files, please copy them completely and replace them in the corresponding locations:
    • old/multi-chat/storage/app/ => new/src/multi-chat/storage/app/
    • old/multi-chat/database/database.sqlite => new/src/multi-chat/database/database.sqlite
    • old/multi-chat/public => new/src/multi-chat/public
    • old/multi-chat/.env => new/src/multi-chat/.env
  3. In addition to these files mentioned in point two, if you have modified or added any other files, please copy them over as well.
  4. If you are using the Windows portable version, please move the following folders or files to their respective locations (since the Python version has changed, there is no need to move the Python folder):
    • old/windows/nginx-1.24.0/ => new/windows/packages/nginx-1.24.0/
    • old/windows/node-v20.11.1-win-x64/ => new/windows/packages/node-v20.11.1-win-x64/
    • old/windows/php-8.1.27-Win32-vs16-x64/ => new/windows/packages/php-8.1.27-Win32-vs16-x64/
    • old/windows/Redis-6.0.20-Windows-x64-msys2/ => new/windows/packages/Redis-6.0.20-Windows-x64-msys2/
    • old/windows/RunHiddenConsole/ => new/windows/packages/RunHiddenConsole/
    • old/windows/composer.phar => new/windows/packages/composer.phar
  5. If you are running on Linux, navigate to new/src/multi-chat/executables/sh/ and run production_update.sh. If you are using the Windows Portable version, run build.bat in new/windows/.
  6. The file update should be completed at this point. You can now check if anything is broken. For the Windows Portable version, please proceed to configure the models according to the tutorial for the new version.

· 2 min read
Yung-Hsiang Hu

Hi, humans! 👋 Welcome to Kuwa! 🤖

Kuwa GenAI OS is an open, free, secure, and privacy-focused Generative-AI orchestrating system,
including user-friendly WebUI for LLMs, and a novel GenAI kernel to support AI-powered applications.

The main features are as follows:

  1. 🌐 Multi-lingual turnkey solution for GenAI development and deployment on Linux and Windows
  2. 💬 Concurrent multi-chat, quoting, full prompt-list import/export/share, and more for users
  3. 🔄 Flexible orchestration of prompts x RAGs x bots x models x hardware/GPUs
  4. 💻 Heterogeneous supports from virtual hosts, laptops, PCs, and edge servers to cloud
  5. 🔓 Open source, allowing developers to contribute and customize the system according to their needs

The Kuwa system was developed with the support of Taiwan's Trustworthy AI Dialogue Engine (TAIDE) project and has been used as a demonstration and development testing platform for the TAIDE project, as well as in several domain-specific applications.

We are a team of students and alumni from the Department of Computer Science and Information Engineering at National University of Kaohsiung, Taiwan, hoping to provide everyone with their own AI development or service platform.
There is still much room for improvement in the Kuwa system, and we sincerely welcome you to join the Kuwa open-source community to participate in this open-source project 🙌. Let’s enter the new generation of GenAI together.

Official website: https://kuwaai.org/
Community: https://kuwaai.org/community
TAIDE: https://en.taide.tw/