SiMa.ai Raises $150 Million to Build Its Next Robot and Drone AI Chip

SiMa.ai has raised $150 million to pursue a two-part pitch to makers of robots, drones and other devices: run AI on its chips, and spend less effort moving applications onto them. The Series C values the company at $1.45 billion. SiMa.ai targets the first half of 2028 for its new hardware’s performance goal.

The software came first

SiMa.ai develops chips and software that let devices such as robots, drones and cameras process AI tasks without sending data back and forth to the cloud. Its challenge is not just to put computing hardware inside those machines. It must also persuade developers to move their applications onto a different platform.

In June, the company launched Palette Neat, an open-source development environment. SiMa.ai says developers can use everyday-language instructions to build applications and map them onto its chips. SiMa.ai says developers can reuse existing code and preserve about 90% of their previous software investment. That percentage is the company’s estimate, not a result guaranteed for every project.

The same June announcement described a production-ready Modalix module designed to fit an Nvidia module form factor without redesigning the board that holds it. SiMa.ai says that existing module can run multiple large language models alongside vision and sensor models under 10 watts. Those are company-supplied specifications for the current hardware, not measurements of the next chip.

Fresh capital for both parts of the pitch

Fidelity Management & Research Company and Amplify co-led the Series C. Alter Venture Partners, Dell Technologies Capital and StepStone Group also participated. The round brings SiMa.ai’s total capital raised to more than $500 million. SiMa.ai says it will use the new funding to scale Palette Neat and build its next-generation Modalix system-on-chip, a processor intended for use inside devices.

The next chip has a 2028 target

SiMa.ai is aiming for 1,000 tera operations per second with its new hardware in the first half of 2028. That is a throughput target, not a measured result from a shipping next-generation chip. The company intends the hardware for drones, humanoid robots, advanced driver-assistance systems and AI-powered car cockpits.

SiMa.ai positions its chips and software as a lower-power, easier-to-deploy alternative to Nvidia’s platform. The financing gives it resources to pursue that claim, but the planned chip’s speed and the ease of moving applications to it remain separate tests. Neither is settled by a future throughput target alone.