Unlocks access to femtoAI’s full-stack AI silicon and compression platform that delivers 10X memory & power savings
SAN BRUNO, Calif., Oct. 7, 2026 /PRNewswire/ — femtoAI, a leading full-stack AI silicon and compression platform, today announced open access to its developer community after a successful beta program. Developers now will have access to femtoAI’s full-stack compression AI platform including the femtoAI Sparse Processing Unit (SPU) chip evaluation kit, compressed models, compiler and Sparsity Studio™ that easily enable development of new AI powered products. femtoAI technology turns patented neural-network sparsity and advanced model compression into real savings, delivering a 10X lower power draw and 10X smaller memory footprint without sacrificing accuracy. Developers can now see what’s possible with femtoAI tools and systems at developer.femto.ai.
“We’ve been working behind the scenes in our developer community with dozens of enterprises as they create new products based on femtoAI’s unique full-stack AI silicon and compression platform. The time to open this experience to more enterprises and developers is now,” said Bill Hoppin, Chief Business Officer, femtoAI. “Customers such as Legato and Marshall gained a competitive advantage by easily developing and deploying using our developer tools and systems. By now opening up the experience for everyone, we can’t wait to see how developers across all industries innovate with the power of our full-stack AI compression platform.”
Key features of the femtoAI developer community include:
- SPU Evaluation Kit: Order your own kit to access the SPU chip
- Self-service development: Access to the full suite of femtoAI tools, compilers and models
- Model flexibility: Bring your own model, use one of femtoAI’s prebuilt models, or develop from scratch with easy-to-use tools and agent-friendly documentation readily available.
- Sparsity Studio™: See actual power and memory savings unlocked by femtoAI sparsity with this measurement tool
- Community Forum: Ongoing open discussion and Q&A with femtoAI experts and developers
“At femtoAI, not only do we deliver first-class sparsity in both our hardware and software tooling and applications,” said Sangnam Choi, VP of Audio & Sensing at femtoAI, “but we also apply advanced model compression techniques that have allowed us to deliver open-source models such as OpenAI Whisper at 10X less memory, and develop custom solutions like local voice commands. Our team continues to unearth and implement emerging techniques and methods, and in turn, we’re able to deliver with an even lower power and memory footprint. We’ll continue to push the possibilities on performance without sacrificing accuracy.”
femtoAI developer community enables faster time-to-market
The femtoAI developer platform has already been instrumental for work across femtoAI’s audio and sensing business with customers such as smart glasses disruptors Legato, and headphone and earbud icon Marshall and innovative companies like NewSound have leveraged femtoAI’s technology to build AI enhanced audio solutions with 10X less memory and 10X less power.
“Integrating advanced AI audio solutions on the femtoAI platform has been a tremendous success for us. The combination of easy-to-use tools, innovative capabilities, and amazing support helped us achieve a level of hearing performance for Legato Frames that was previously impossible in the form factor on a remarkable timeline,” Mehul Trivedi, CEO at Legato Hearing said. “We’ve loved working with the femtoAI team and are excited to continue delivering industry-leading products with them. It’s an extraordinarily valuable partnership for Legato, both now and as we grow.”
Join the Community
Meet the femtoAI team on Wednesday October 7 for an SF Tech Week developer happy hour at Barebottle Brewing Company’s beer garden located at Salesforce Park. You’ll enjoy drinks, light bites, and an exclusive walk-through of the developer community from femtoAI experts. RSVP today to join us.
To join our developer community and access the femtoAI platform, visit developer.femto.ai.
About femtoAI
femtoAI is a leading full-stack AI silicon and compression platform that turns patented neural-network sparsity into real savings for enterprises. Headquartered in Silicon Valley, femtoAI was founded in 2018 by researchers from the Brains in Silicon Lab at Stanford University. femtoAI’s full-stack AI silicon and compression platform enables low-latency, 10X lower power draw and 10X smaller memory footprint for customers like Marshall, NewSound and Legato. From wearables and household appliances to smart glasses and robotics, femtoAI brings the power of AI to every device.
SOURCE femtoAI
