January 17, 2024
News
![](https://data.embeddedcomputing.com/uploads/resize/1256/756/external/data.embeddedcomputing.com/uploads/articles/primary_images/1705515407.jpg)
Tel Aviv, Israel. Deci is teaming up with Qualcomm Technologies, Inc. to pioneer Generative Artificial Intelligence (AI) models designed for the Qualcomm Cloud AI 100. The companies hope to make AI available for numerous AI-powered applications and democratize AI technology for all. Deci and Qualcomm Technologies will introduce two models, the DeciCoder-6B and DeciDiffusion 2.0, developed on Deci’s Neural Architecture Search Technology, AutoNAC.
“Together with Qualcomm Technologies we are pushing the boundaries of what’s possible in AI efficiency and performance” said Yonatan Geifman, CEO and co-founder of Deci. “Our joint efforts streamline the deployment of advanced AI models on Qualcomm Technologies’ hardware, making AI more accessible and cost-effective, and economically viable for a wider range of applications. Our work together is a testament to our vision of making the transformational power of generative AI available to all.”
The DeciCoder-6B is a 6 billion limit model for code generation supporting C, C#, C++, GO, RAST, Python, Java, and JavaScript. According to the press release, the solution surpasses other models including CodeGen2.5-7B, StarCoder-7B, and CodeLlama-7B. The DeciCoder-6B is said to have 19 times more data compared to comparable models when utilized on Qualcomm’s Cloud AI 100
DeciDiffusion 2.0 boasts 732 million parameter text-to-image that is reported to best Stable Diffusion v1.5 with operation at 2.6 times the speed for quality imaging. Both were published under Apache-2.0 and CreativeML Open RAIL++-M Licenses.
For more information, visit Deci and Qualcomm Technologies, Inc.
And for more information on the democratization of edge AI don’t miss our Embedded Insiders Podcast.