Chuxin Partners Capital is a venture capital investment firm.

Chuxin Partners Capital

Chuxin Capital is an early-stage fund that actively deploys emerging formats centered on technology and consumption upgrades. It specializes in early-stage investments into companies within the enterprise services, technology, entertainment, and consumer products and services industries. Chuxin Capital was established in 2015 and is headquartered in Beijing, China.

Chuxin Capital

HongShan is a VC firm focused on seed-stage, mid-stage, late-stage, and growth investments in the fintech sector. HongShan has offices in Hong Kong, Beijing, Shanghai, Guangzhou, Hangzhou, Suzhou, and Shenzhen. Over the past nine years, Sequoia Capital China has an impressive and diverse portfolio comprising about 300 dynamic companies that deliver high returns on investment using differentiated technologies and innovative business models, including Alibaba, VIPshop.com, Sina.com, Didi, JD.com, Ourpalm, Qihoo 360, Jumei, Momo, SINNET, Dianping.com, Meituan, Meilishuo, Toutiao, AutoNavi, Ganji.com, DJI, VanceInfo, Noah Private Wealth Management, Wanda Cinemas, Alibaba Pictures, Plateno Hotels Group, Deppon Logistics, ZTO Express, Beta Pharma, Snibe Diagnostic, BGI, WEGO, and Yuwell Medical.

HongShan

CoCoPIE

CoCoPIE is a startup developing a platform to optimize AI models for edge devices. CoCoPIE reduces the expected time to market in order to meet if not exceed demand for applications.

CoCoPIE is a software company that develops AI optimization and acceleration technology for mobile and IoT devices. Founded in 2021, the company's core technology is compression-compilation co-design, which allows for the efficient deployment of AI models on end devices while maintaining accuracy. CoCoPIE's flagship product, XGen, is an AI acceleration toolchain that co-optimizes deep neural network (DNN) models and code, making AI models significantly smaller and faster for use on devices. XGen supports integration with Android and iOS applications and can be deployed on individual computers or clusters for scalable use.

Unlocking AI for End Devices: CoCoPIE Launches Industry-First DNN Model-Code Co-Optimization Tool (XGen) for Device AI

CoCoPIE Launches Industry-First, Cross-Platform Real-Time Super-Resolution (CRSR) Technology for Upscaling Low Res Videos on the Fly

Dr Xipeng Shen of CoCoPIE: Five Things You Need To Create A Highly Successful Startup

CoCoPIE Launches CRSR Technology

CoCoPIE Launches DNN Model-Code Co-Optimization Tool for Device AI

CoCoPIE launches XGen

AI tool optimizes models, code for end devices

Reinventing how computers think based on biological neural networks

Neuromorphic Computing

CoCoPIE is a software company that develops AI optimization and acceleration technology for mobile and IoT devices. Founded in 2021, the company's core technology is compression-compilation co-design, which allows for the efficient deployment of AI models on end devices while maintaining accuracy. CoCoPIE's flagship product, XGen, is an AI acceleration toolchain that co-optimizes deep neural network (DNN) models and code, making AI models significantly smaller and faster for use on devices. XGen supports integration with Android and iOS applications and can be deployed on individual computers or clusters for scalable use.
The company also offers Cross-platform Real-time Super-Resolution (CRSR) technology, which transforms low-resolution videos into high-resolution in real-time on various handheld devices. CRSR improves video super-resolution speed from several frames per second to over 30 frames per second, enabling real-time super-resolution on smartphones, including low-end models. This technology aims to enhance user experience and save network bandwidth for video streaming providers.

MemComputing, founded in San Diego, has developed a novel computing technology aimed at solving complex optimization problems across various industries. The company's proprietary MemCPU Platform combines processing and memory to deliver high-performance computing capabilities for applications in optimization, Big Data analytics, and machine learning. MemComputing's approach is based on a physics-inspired computing paradigm that utilizes non-quantum dynamical systems with memory to process information.
The technology behind MemComputing was invented by Fabio Lorenzo Traversa and Massimiliano (Max) Di Ventra, who sought to design a computer from a physics perspective. Their innovation led to the creation of digital memcomputing machines (DMMs), which are designed to solve combinatorial optimization problems efficiently. DMMs can be realized through electrical circuits or simulated on traditional computers.
MemComputing's solutions have demonstrated significant advantages in solving hard problems. For instance, the company reported finding feasible solutions within minutes for a problem from the MIPLIB 2010 library whose feasibility had remained unknown for eight years. In another case, MemComputing's technology solved a problem in 1.5 hours that would have taken the world's fastest supercomputers significantly longer to solve.
The company's first product, the MemCPU Extreme Performance Computing (XPC) SaaS, is being used by companies in the Global 2000 to address large-scale optimization challenges. MemComputing's approach has shown promising results in various applications, including prime factorization, Boolean satisfiability problems, and restricted Boltzmann machine training.

MemComputing

MemryX Inc. is a fabless semiconductor company that specializes in designing edge AI processing solutions. Founded in 2019, the company focuses on accelerating artificial intelligence (AI) processing for edge devices through its proprietary compute-at-memory technology and dataflow architecture. MemryX's flagship product, the MX3 AI Accelerator, is designed to be an efficient and easy-to-use solution for edge AI applications. The MX3 can be integrated into various edge computing systems, offering scalable AI capabilities for industries such as automotive, robotics, IoT, smart vision systems, and edge computing.
The MX3 AI Accelerator features MemryX Compute Engines (MCE) that are tightly coupled with at-memory computing, creating a native dataflow architecture that utilizes up to 70% of the chip with just one click. This design allows developers to compile any trained AI model with a single click, achieving optimized model performance with high utilization (>50%) and full accuracy within minutes. The MX3 supports multiple AI frameworks, including TensorFlow, TensorFlow Lite, ONNX, PyTorch, and Keras, and can run hundreds of unaltered AI models efficiently.
In terms of performance, the MX3 offers more than five TFLOPs of computing power while consuming approximately one watt of power. It supports BFloat16 activations and can handle weights of 4, 8, and 16 bits. The chip can store about 10 million parameters on-die and interfaces with host systems through PCIe Gen 3 I/O and/or USB 2.0/3.x. MemryX offers the MX3 in various form factors, including bare die, single-die package, and modules (mini PCIe or M.2) with one or more MX3 chips.
As of December 2023, MemryX announced the availability of production-level silicon for its MX3 AI Accelerator, with volume production expected to start in 2023. The company has showcased various demonstrations of the MX3's capabilities, including multi-stream AI-enabled video management systems, AI-powered PCs, threat and weapons detection for public spaces, and smart city traffic management solutions.
Key customers and partnerships
MemryX has formed partnerships with several companies to showcase and integrate its AI acceleration technology. In February 2023, MemryX announced a partnership with Edge Impulse, an AI-algorithm software provider, to streamline the development, testing, and deployment of AI models on MemryX's AI accelerator chips. This collaboration aims to simplify the process of creating and implementing AI solutions for edge devices.
The company has also worked with Cachengo to develop an intelligent edge platform for real-time threat detection in public spaces. Additionally, MemryX has partnered with Fracctal to demonstrate smart city traffic management capabilities powered by the MX3 AI Accelerator.

MemryX

Nepes Corporation is a South Korean company specializing in Outsourced Semiconductor Assembly and Test (OSAT) services. The company provides a comprehensive range of semiconductor packaging design and manufacturing services to global electronics industry clients. Nepes offers various packaging solutions, including wafer level packaging, fan-out wafer level packages, and panel level packaging. In 2020, Nepes acquired factory assets from Deca Technologies, establishing itself as a key player in advanced packaging technologies. The company has implemented Deca's M-Series fan-out technology and Adaptive Patterning real-time EDA ecosystem, enabling it to build the industry's first 600mm large panel M-Series factory in South Korea.
Nepes has expanded its IC packaging capabilities to address the challenges of developing advanced 3D IC packages. The company leverages a range of industry-leading solutions from Siemens EDA, including the Calibre nmPlatform tool, HyperLynx software for electrical rule checking, Calibre 3DSTACK software, Xpedition Substrate Integrator software, and Xpedition Package Designer software. These technologies enable Nepes to provide fast and reliable design services, including 2.5D/3D-based chiplet designs for its growing base of global IC customers.
Key customers and partnerships
Nepes has established partnerships with several prominent companies in the semiconductor industry. The company entered into a technology transfer and licensing agreement with Deca Technologies in 2020, enabling it to implement advanced packaging technologies. Nepes also collaborates with SAPEON Korea, providing them with comprehensive semiconductor packaging design and manufacturing services. Additionally, Nepes has extended its relationship with Siemens Digital Industries Software, leveraging their EDA technologies to enhance its packaging design capabilities and serve mutual customers more effectively.

Nepes

Symetrix Corporation specializes in developing technology for the semiconductor industry. The company's innovations find applications across various sectors, including communications, data acquisition, and measurement devices. Symetrix's technology serves as a crucial component in the advancement of semiconductor capabilities, contributing to the evolution of diverse electronic systems. The company's focus on memory technology suggests a significant role in enhancing data storage and processing capabilities within the semiconductor field.

Symetrix

Syntiant is a deep learning chip technology company founded in 2017 and headquartered in Irvine, California. The company specializes in developing ultra-low-power, high-performance deep neural network processors called Neural Decision Processors (NDPs) for edge AI applications. Syntiant's technology combines deep learning with semiconductor design to enable AI functionality in battery-powered devices without relying on cloud connections.
Syntiant's NDPs are designed to execute deep neural networks natively on various topologies, including Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and fully connected networks. The company's chips typically offer 100x efficiency improvement and a 10x increase in throughput compared to traditional low-power microcontroller-based solutions. This allows for larger networks at significantly lower power consumption, extending battery life and reducing infrastructure costs.
The company's product lineup includes the NDP100, NDP101, NDP200, and NDP115 processors. The NDP200, built on the Syntiant Core 2 architecture, provides 25 times the tensor throughput of the Syntiant Core 1 found in the NDP100. The latest NDP115, introduced at CES 2023, is designed for ultra-power and size-constrained applications, capable of running speech inferences at 280 microwatts.
Syntiant has achieved significant milestones in its short history. By January 2021, the company had shipped over 10 million of its NDP100 and NDP101 processors. This number increased to more than 20 million processors shipped by March 2022. As of May 2023, Syntiant has expanded its offerings to include hardware-agnostic deep learning computer vision models that can be deployed on various platforms, including the Renesas RZ/V2L microprocessor.
Key customers and partnerships
Syntiant's technology has found applications across various industries and devices. The company's Neural Decision Processors are being used in earbuds, wearables, smartphones, smart speakers, laptops, vehicles, and other IoT consumer and industrial use cases.
Syntiant has established partnerships with several prominent companies in the tech industry. In July 2021, Syntiant announced a collaboration with Renesas Electronics Corporation for a voice-controlled multimodal solution combining Renesas' RZ/V Series MPUs and Syntiant's NDP120 Neural Decision Processor. This partnership aims to enable low-power contactless operation for image processing in vision AI-based IoT and edge systems.
The company has also worked with Infineon Technologies, signing a smart microphone reference design agreement in 2018. This collaboration aimed to complement Syntiant's NDPs with Infineon's high-performance microphone technology.

Syntiant

Develop neural network hardware, including chipsets and circuits

Software companies that offer tools that can be used to develop neuromorphic AI algorithms and solutions

Develop software solutions to can be used to design neural networking tools, or offer neural networking solutions directly to users

Software companies that apply neuromorphic computing for visual AI tools and applications

Companies that develop and offer solutions that incorporate Neuromorphic-based AI technology

AI/ML enables neuromorphic computing to be applied to artificial neural networks and provide functional outcomes

CoCoPIE

Sources

About EDGE

Contact Us

Resources

Area of Focus

Technology Research Made Effortless