SiMa.ai
Component
San Jose edge-AI silicon company selling the MLSoC and Palette software for on-device physical-AI inference.
SiMa.ai builds purpose-built edge-AI silicon and software: the Machine Learning System-on-Chip (MLSoC) combined with its Palette agentic software platform, delivering high-performance, low-power inference for computer vision, conversational AI and multimodal workloads directly on devices - robots, vehicles, drones and industrial/healthcare systems - without cloud dependency. Founded in 2018 by former Xilinx and Groq executive Krishna Rangasayee, SiMa.ai develops machine-learning IP, chiplets and SoCs, and positions itself for the 'physical AI' wave across humanoids, automotive and drones. The company reported unicorn status at a $1.45B valuation in 2026. (Company-reported figures.)
Research mapping
Physical AI / Edge Compute & Control Silicon (L5) / Component
Evidence boundary
Original analysis and independent corroboration remain under development. Commercial scale and durable technical advantage remain under review.
What makes it defensible
Software-first, purpose-built MLSoCs (chip plus Palette software) tuned for edge inference efficiency - a full-stack edge platform versus competitors repurposing cloud silicon.
Position in the stack: Edge-AI silicon and software vendor supplying SoCs/chiplets and a deployment platform to device and robot OEMs.
Who buys it
Founding team
Krishna Rangasayee — Founder & CEO — 18 years at Xilinx rising to SVP/GM and EVP of global sales (grew business to $2.5B); previously COO of AI-inference chipmaker Groq; holds 25+ international patents.
SourceFinancing detail
- Last round
- Series C
Investors: Fidelity, Amplify, Dell Technologies Capital, Micron Technology
Mapped funds and investors
Direct relationships supported by published evidence.
Reported holdings as-of
No sourced relationships mapped yet.
Disclosed investment history
No sourced relationships mapped yet.
Showing up to four relationships in each group. ETF weights describe the fund’s portfolio, not ownership of this company. Private investment history does not establish current ownership.
Evidence & media
-
SiMa.ai Reaches $1.45B Valuation with $500 Million in Total Funding to Scale Physical AI
What this source supports: $150M Series C at $1.45B; co-led by Fidelity and Amplify; total $500M; new investors AllianceBernstein, Baron Capital, J.P. Morgan, State of Michigan
What it does not establish: Press release; revenue and customer names undisclosed
Source -
Physical AI Chipmaker SiMa.ai Lands $150M Series C, Hits $1.45B Value
What this source supports: Modalix MLSoC: 50 TOPS multimodal at sub-10W; pin-compatible with NVIDIA Jetson Orin NX/Nano for drop-in upgrades
What it does not establish: Specs from vendor documentation; no independent benchmark vs NVIDIA
Source -
SiMa.ai Launches Palette Neat, the Industry's First Agentic Environment for Physical AI
What this source supports: Company release: Palette Neat is an agentic (AI-agent-driven) application development environment for Physical AI on SiMa.ai's Modalix MLSoC, claimed to cut application development from months to days.
What it does not establish: 'Industry first' is a company claim; no adoption, developer-count or customer metrics.
Source -
SiMa.ai and Wobot.ai Partner to Bring Reasoning-Based Video Analytics for Physical AI
What this source supports: BUSINESS WIRE (SAN JOSE, Oct. 5, 2026): Wobot.ai's deployment-ready, reasoning-based video analytics now runs on SiMa.ai's Modalix MLSoC PCIe card and SoM, keeping video on-premises; Wobot analytics cited as deployed across QSR, retail, hospitality and manufacturing.
What it does not establish: No unit volumes, revenue or joint-customer names; "field-proven" deployment breadth is Wobot's own framing.
Source -
SiMa.ai and Wobot.ai Partner to Bring Reasoning-Based Video Analytics for Physical AI
What this source supports: Edge AI and Vision Alliance (October 5, 2026): independent trade reproduction corroborating the Wobot.ai-on-Modalix partnership and the edge/privacy rationale (video stays on-site for EU/Japan data-residency requirements).
What it does not establish: Reproduces the release; no independent adoption data.
Source
Timeline
-
partnership
Wobot.ai partnership — reasoning-based video analytics runs on Modalix MLSoC PCIe card and SoM, on-premises at the edge
Source -
Product
Launches Palette Neat — an agentic development environment that collapses physical-AI application development from months to days on Modalix MLSoC
Source -
Financing
Series C — $355,000,000 total disclosed
Source