Verkor Launches Industry's First TurboQuant LLM Inference Accelerator Silicon IP
In recent news, a new accelerator chip called VerTQ has been introduced, which incorporates Google's TurboQuant algorithm. This innovative chip has successfully reduced the KV cache memory usage of Large Language Models by an impressive factor of 4.3x. Remarkably, despite this reduction, it has been able to maintain or even improve the performance of the models it supports. Notably, VerTQ was autonomously developed by Conductor 2.0, Verkor's secondary project.
Source: Unusual Whales The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.
Post your comment
No Comments