SiMa.ai, a system-on-chips (SoC) manufacturer for Edge computing, has introduced MLSoC Modalix, a new product family for multimodal Edge AI.
The MLSoC Modalix supports various AI models, including convolutional neural networks (CNNs), transformers, large language models (LLMs), language models with memory (LMMs), and GenAI at the edge. The company claims it delivers over 10x the performance per watt compared to alternatives.
The chip also includes SiMa.ai's Palette software with a patented compiler for efficient data management.
Analyst QuickTake: The release of MLSoC Modalix aligns with SiMa.ai’s focus on enhancing performance for Edge AI. The chip enables end-to-end Gen AI pipelines on a single chip, reducing latency, power consumption, and cloud reliance. This improves real-time performance, enhances data security through local processing, and simplifies AI development by minimizing hardware dependencies.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.