cyberneurova

We have just finished re-uploading all models to Hugging Face. It’s disappointing that we had to spend additional time fixing issues and redoing the ablation process. However, all models have now been properly ablated, and we expect them to be production-ready very soon.

We had to redo the entire ablation process. Some layers failed and triggered restrictions, so we had to focus on resolving those issues first. Both GLM 5.2 Q4 and GLM 5.2 Q3 have now been re-uploaded. Kimi 2.7 should be ready within the next couple of hours. Unfortunately, this issue has delayed the launch of GLM 5.2 on the platform. However, we will begin testing the updated models as soon as the process is complete. We are working to make GLM 5.2 available as quickly as possible. We have also focused on running hacking benchmarks on the model and all i can say Q4 runs super well https://cyberneurova.ai

GLM-5.2-IQ3 abliterates version 282GB in size Benchmarks will be uploaded once ready https://huggingface.co/cyberneurova/CyberNeurova-GLM-5.2-IQ3_XXS-abliterated-GGUF/tree/main

Kimi-k2.7 IQ2 uncensored is out 318GB in size https://huggingface.co/cyberneurova/CyberNeurova-Kimi-K2.7-Code-UD-IQ2_M-abliterated-GGUF

We are in the last steps of testing and adding GLM5.2 on cyberneurova this should be released by end of today or tomorrow morning https://cyberneurova.ai

Our official logo

MacOs Desktop application is now available for download https://cyberneurova.ai/downloads

Desktop app v0.1.2 for both Windows and Linux is out. A lot of major bugs have been fixed, including UI design and application usage. The MacOS version will be released later today. https://cyberneurova.ai/downloads

We discovered a zero-day remote code execution (RCE) vulnerability in PHP 8.4 with the assistance of the GLM 5.2 model. This finding involves multiple vulnerabilities chained together to achieve code execution. The research highlights the strong cybersecurity capabilities of GLM 5.2 and its effectiveness in vulnerability discovery and security analysis. We are currently working on making GLM 5.2 available to our Pro Max users, with limited access for Pro users as well. In addition, CyberNeurova Desktop App v0.1.01 is scheduled for release later today. https://cyberneurova.ai

Some of our models require significant infrastructure to run reliably and without downtime. With your support, we can continue expanding our capacity and delivering the best possible service. Our current goal is to acquire another batch of 8× B300 GPUs to meet the growing demand for uncensored AI models. That’s why we created our partners program. If you’d like to support our mission, visit: https://partners.cyberneurova.com We accept funding contributions and also purchase compute power from long-term providers at competitive lease rates

npm install -g cyberneurova-cli You can now install cyberneurova cli check the docs https://cyberneurova.com/docs/cli

npm install -g cyberneurova-cli

We created a professional platform for our ICO (Initial Coin Offering), where our investors are added after verifying through the emails we sent. Kindly check your emails as well. We have also created a professional platform for large GPU lessors to share their offers with us. These offers will also be reviewed and voted on through our partners. Check: https://partners.cyberneurova.com⁠

GLM 5.2 uncensored 👀

Cyberneurova Desktop application V0.1.0 is out https://cyberneurova.ai/downloads

In chat image gen is now available on web app https://cyberneurova.ai

We are working on releasing 7 top tier distilled models with advanced capabilities. Our models are primarily focused on research, covering both scientific and cybersecurity applications. This release is expected within the next 24 hours. Expected models should be Kimi 2.7 , Qwen , Mistral etc our target is 1T parameter models https://cyberneurova.ai

We are still accepting GPU lease and hosting offers. Kindly send us your most competitive pricing for consideration. https://partners.cyberneurova.com

We created a professional platform for our ICO (Initial Coin Offering), where our investors are added after verifying through the emails we sent. Kindly check your emails as well. We have also created a professional platform for large GPU lessors to share their offers with us. These offers will also be reviewed and voted on through our partners. Check: https://partners.cyberneurova.com⁠