News Nvidia reveals Large Plant Model to zap weeds using laser beams from tractors fitted with nvidia GPUs

marees

Platinum Member
Apr 28, 2024
2,268
2,891
96
The shovel maker, constantly keeps digging up, new use cases for his shovels


LaserWeeder packs two dozen Nvidia GPUs and lasers to zap your weed problem, vaporizes ‘600,000 weeds per hour' with sub-millimeter precision — instant laser death for pesky weeds​

News
By Mark Tyson published August 28, 2025
Nvidia and Carbon Robotics’ tractor-pulled device is designed to kill large areas of weeds

Carbon Robotics LaserWeeder G2 packs in 24 high power lasers and 24 Nvidia GPUs to facilitate its claimed “10,000 weeds per minute” zapping performance.

That's equivalent to 600,000 weeds per hour, or 167 weeds per second.

The ‘Large Plant Model’ weed identification system, ensures your crops don’t fall to friendly fire.

The robotics device is also a massive labor saver, outperforming “a hand crew of 75 people,” according to the product pages for the G2 600 model.


https://www.tomshardware.com/tech-i...eath-for-pesky-weeds#xenforo-comments-3885403



Carbon Robotics’ story began not in a lab, but over lunch in Idaho. Founder Paul Mikesell asked a farmer about his biggest challenges. The answer was simple — and ancient: weed control.

That conversation led to an audacious solution. The company’s LaserWeeder G2 spans 20 feet and carries 12 modules, each powered by two NVIDIA GPUs — 24 in total. This compute muscle lets it identify and incinerate up to 10,000 weeds per minute, without chemicals.


Every LaserWeeder doubles as a data-gathering machine. Images from the field feed into a custom labeling tool, building what the team believes is the world’s largest labeled agricultural image dataset: over 65 million images, fueling the company’s “large plant model,” a play on the large language models that power chatbots like ChatGPT.


When GPUs finish their tour on the farm, they head to Carbon’s “retirement facility” — a Seattle-based data center — where they’re used to train the next generation of models. Optimized with NVIDIA TensorRT, the resulting foundational model works across all crops, creating a durable competitive edge.



https://blogs.nvidia.com/blog/2-million-robotics-developers/#carbon-robotics
 
  • Like
Reactions: Elfear

DaaQ

Platinum Member
Dec 8, 2018
2,038
1,457
136
The LPM can differentiate between weed & plant. Can Monsanto do the same 😉
Its green it grows, what else it need to know?

For the combine to not zap the corn ears?
 

coercitiv

Diamond Member
Jan 24, 2014
7,485
17,887
136
Nvidia should stick to GPUs for gamers.
Once you're done zapping the crops you can host a LAN party right there, in the field.

Certainly better than Chems. If it works at a reasonable cost.
I'm skeptical about the cost in terms of hardware, energy consumption & pollution and also it's ability to fully replace chemicals. It's efficiency likely diminishes once the crop starts to grow. Then again, for organic crops the math might check out.

They claim 1 to 3 years to recoup the investment, but I don't see any pricing on their site. They also claim up to 50% yield improvements for traditional crops due to plants growing better when subjected to less chemical treatments. Overall we have to trust them that once all the savings add up, the device pays for itself relative to chemical + additional manual labor that is not needed for the laser solution.
 
  • Like
Reactions: marees