SiMa.ai Raised $150 Million to Scale AI Silicon
The San Jose startup targets 1,000 TOPS performance for its next-generation machine learning chips by 2028.
Updated on Oct. 5, 2026 in Semiconductors
Live Poll
Do you believe current high valuations for physical AI hardware companies are justified?
San Jose-based SiMa.ai has raised $150 million in an oversubscribed Series C funding round. The investment brings the company's total capital raised to roughly $500 million, supporting a post-money valuation of $1.45 billion.
Why it matters
The capital infusion accelerates SiMa.ai's development of specialized hardware, including machine-learning IP and systems-on-chip, to meet growing demand for local AI processing power. The company is positioning its hardware to support intensive computation workloads expected in the next few years.
SiMa.ai is targeting a compute performance of 1,000 dense TOPS (trillion operations per second) for its upcoming physical AI silicon. This performance goal serves as the benchmark for its next phase of hardware development.
The players
SiMa.ai
A San Jose-based developer of purpose-built machine learning hardware, chiplets, and systems-on-chip.
Fidelity Management & Research Company
An investment firm that co-led the Series C round, known for managing large-scale mutual fund capital.
Amplify
An investment entity that co-led the financing round for the chip startup.
The details
The company intends to use the capital to scale its Palette Neat software environment—a development platform designed to streamline the deployment of machine learning models onto custom silicon. Simultaneously, the funding supports the design of new chiplets and systems-on-chip—integrated circuits that incorporate multiple components of a computer onto a single substrate. These hardware efforts are directed toward delivering high-density computing power for edge and industrial AI applications.
Timeline
October 5, 2026: SiMa.ai announced the $150 million Series C funding round.
First half of 2028: Target release window for the new physical AI silicon hardware.
The Tech Race
SiMa.ai is positioning its custom silicon development against the incumbent providers of large-scale AI training and inference hardware. The firm's focus on chiplets and systems-on-chip targets a specialized market segment aiming to provide high compute density for specialized AI workloads.
This development impacts enterprise and industrial users who require high-performance local AI processing, though consumer-grade hardware availability remains contingent on the 2028 hardware roadmap. Organizations and developers should watch for updates to the Palette Neat software environment as it scales toward these future silicon releases.
The takeaway
SiMa.ai is betting heavily on custom, dense silicon to solve future machine-learning bottlenecks for industrial users. Investors and partners should track the firm's progress toward its 1,000 TOPS performance milestone scheduled for the first half of 2028.
What happens next
The company is slated to deliver its next-generation physical AI silicon, which aims to reach 1,000 dense TOPS, in the first half of 2028.
Further reading
For broader trends in hardware development, explore the latest updates in Semiconductors.
Source note: This article includes information reported by Hedgeco.
Live Poll
Do you believe current high valuations for physical AI hardware companies are justified?
