About deepsilicon
Deepsilicon set out to build software and hardware for ternary transformer models, neural networks whose weights take only the values minus one, zero or one. Such weights fit in two bits instead of sixteen, which the founders estimate gives close to eight times compression, and multiplications reduce to additions and subtractions. The company argues that current chips are not designed for such low precision. It joined the Y Combinator Summer 2024 batch, was founded in 2024 in San Francisco with a team of two, and Y Combinator lists its status as inactive. No shipped chip was found.
deepsilicon was founded in 2024 and is based in San Francisco, CA, United States. It is listed under AI Accelerators, ASICs. Reported funding is YC S24.
Compute Technologies
- AI Accelerators
- ASICs
What deepsilicon works on
AI Accelerators
An AI accelerator is a chip, or a part of a chip, designed to run machine learning faster or with less energy than a general-purpose processor. The category includes GPUs, TPUs, neural processing units (NPUs) and chips from many startups.
ASICs
An ASIC, or application-specific integrated circuit, is a chip designed to do one job or a narrow set of jobs. Its circuits are fixed when the chip is made. A CPU can run any program; an ASIC does what it was designed for and nothing else.
Also in United States
Ouster, Cadence Design Systems, Uhnder, Empower Semiconductor, Duality Technologies
