The global market forecast for AI accelerator chips, which is expected to grow 35 percent annually from $8 billion in 2019 to $70 billion in 2026 according to research firm Global Market Insights, underscores the massive amount of cash and growth opportunities that are at stake in the AI chip war, says the publication. Its choices for the 10 "coolest" AI startups of 2020 are based on recent milestones the companies have made, such as funding rounds, product launches, or performance records:
- Blaize - Its Graph Streaming Processor, offered as the first to run multiple artificial intelligence models and workflows on a single system at the same time, is claimed to overcome barriers in AI processing cost and size, providing 10 to 100 times greater efficiency over existing solutions.
- Cerebras Systems - The computer systems company's Wafer Scale Engine processor - claimed to be the largest chip ever built, measuring at 1.2 trillion transistors and packing 400,000 compute cores - enables AI research at previously-impossible speeds and scale.
- Flex Logix Technologies - The company says its InferX X1 is the fastest and most efficient edge AI inference accelerator.
- Graphcore - Its Intelligence Processing Unit chip, says the company, is the first processor designed from the ground up for machine intelligence.
- Hailo - The company offers what it says is the world’s top performing AI processor for edge devices, the Hailo-8 deep learning chip.
- Kneron - This on-device-oriented AI solution provider develops AI chips for edge devices that can adapt to audio and visual recognition applications on the fly.
- LeapMind - This processor IP company offers a design for an ultra-low-power AI inference accelerator for ASIC and FPGA circuits that it says can run AI models in small, 1- to 2-bit data formats at nearly the same accuracy as 8-bit formats.
- SambaNova Systems - The company says its software-defined-hardware approach using its reconfigurable dataflow architecture allows applications to take the lead in driving how hardware is optimized to accelerate performance in data centers and at the edge.
- SiMa.ai - The company says its Machine Learning System-on-Chip (MLSoC), which is designed to be environmentally friendly and efficient, is the first chip to combine high performance, low power, and hardware security for machine-learning inference.
- Tenstorrent - The company's Grayskull AI processor is offered as being able to unlock new levels of deep learning performance with the industry’s first conditional execution architecture that dynamically eliminates unnecessary computation.
For more, see the article at CRN: "The 10 Coolest AI Chip Startups Of 2020."
Giant trillion transistor chip built for AI
AI chip startup emerges offering next-gen computing architecture
'Future-proof' chip design aims to disrupt AI space
Fastest AI inference chip for edge systems announced
AI processor is 'world's most advanced'
Intel invests in 'groundbreaking' AI chip architecture startup