By using this site, you agree to the Privacy Policy and Terms of Use.
Accept
Techticia.comTechticia.comTechticia.com
  • Home
  • Recent
  • Jobs
  • Categories
    • News
    • AI
    • Apple
    • Gadgets
    • Startups
    • Wearables
  • Pages
    • Privacy Policy
    • About Us
    • Contact Us
    • Terms and Conditions
Search
Emerging Tech
  • Wearables
Business & Policy
  • Startups
Mobility & Culture
  • Contact
  • Blog
  • Privacy Policy
  • Terms of Use
© 2026 Techticia.com. All Rights Reserved.
Reading: Tiny LLM Smartphone: PrismML Shrinks AI for Your Pocket
Share
Sign In
Notification Show More
Font ResizerAa
Techticia.comTechticia.com
Font ResizerAa
Search
  • Home
  • Recent
  • Jobs
  • Categories
    • News
    • AI
    • Apple
    • Gadgets
    • Startups
    • Wearables
  • Pages
    • Privacy Policy
    • About Us
    • Contact Us
    • Terms and Conditions
Have an existing account? Sign In
Follow US
  • Contact
  • Blog
  • Privacy Policy
  • Terms of Use
© 2026 Techticia.com. All Rights Reserved.
Techticia.com > Blogs > AI > Tiny LLM Smartphone: PrismML Shrinks AI for Your Pocket
AIGadgets

Tiny LLM Smartphone: PrismML Shrinks AI for Your Pocket

Matilda Lucia, Techticia.com author avatar
Last updated: September 19, 2026 4:20 am
Matilda
Share
Tiny LLM Smartphone: PrismML Shrinks AI for Your Pocket
Tiny LLM Smartphone: PrismML Shrinks AI for Your Pocket
SHARE

Tiny LLM Smartphone: The Future of On-Device AI

PrismML believes powerful reasoning models don’t need to be massive. The Caltech-founded AI lab is developing tiny LLM smartphone technology that could change how we all use AI — by running advanced models directly on your PC or phone instead of in the cloud.

Contents
  • Tiny LLM Smartphone: The Future of On-Device AI
    • Who Is Behind PrismML?
  • What Makes Bonsai 2 Different?
    • How Does the Compression Work?
  • Can Tiny LLMs Match Full-Size Models?
    • What’s Next for PrismML?
  • Why On-Device AI Matters

On Thursday, PrismML released Bonsai 2 27B, its latest compressed model. It shrinks Qwen3.8 27B, a widely used open source model from Alibaba, down to just 5.9 GB. That’s small enough to fit on a PC and possibly a high-end smartphone — a 9x to 10x reduction in memory versus the original.

Who Is Behind PrismML?

PrismML was founded by a group of Caltech researchers and is led by CEO Babak Hassibi, a Caltech professor and expert in compression technologies. The startup also counts Ion Stoica as an adviser. Stoica is a co-founder of Databricks and director of Berkeley’s Sky Computing Lab, which has birthed technologies and startups from Letta to SGLang.

The company is backed by investors Khosla Ventures, Cerberus Capital, and Caltech. It has raised a $22.25 million seed round.

What Makes Bonsai 2 Different?

PrismML is not the only company working on LLM compression. Multiverse Computing, founded by a professor from Spain’s Donostia International Physics Center, is another player — and has raised significant funding.

But Hassibi says PrismML’s compression tech is unique because its models lose virtually no performance compared with the originals. Bonsai 2 matches 98% of Qwen’s aggregate benchmark scores, up from the first Bonsai released in March, which matched 95%.

That original model has already been downloaded over 11 million times, and PrismML’s even smaller models have been downloaded another 2.6 million times, according to the company.

How Does the Compression Work?

PrismML achieves this by shrinking the “weights” that make up a model — the information a model learns and stores during training. Normally, each weight requires 16 bits. PrismML’s approach, called “ternary” weights, simplifies that down to three values: +1, −1, or 0.

With far smaller values to store for each weight, the model takes up dramatically less space. For a deeper dive on the compression technique, the project’s Hugging Face page offers more detail.

Can Tiny LLMs Match Full-Size Models?

Whether PrismML could ever reach 100% benchmark performance parity remains to be seen. Compression will likely always have some impact, Hassibi says.

Still, perfect benchmark parity is fairly academic. LLMs are not so accurate in their uncompressed form — and benchmarks not so perfectly reflective of actual tasks — that a 2% degradation would likely meaningfully affect how a model performs in real use. The surrounding software, or the harness a model runs inside of, also matters a lot when it comes to accuracy.

What’s Next for PrismML?

The startup’s next goal is to apply this compression technique to even bigger models. “The next models that we will release, hopefully in the next couple of months, will be in the several-hundred-billion-parameter range, and I expect it will be easier to retain the intelligence there,” Hassibi told TechCrunch.

As model size grows, he added, “There is more room to be able to compress them without losing the intelligence. So I would just say, as a general trend, for larger models, it’s easier to get to 100%.”

Why On-Device AI Matters

Stoica says he’s excited for this tech because it makes it possible for advanced models to run on users’ devices. “You are going to have intelligence at your fingertips, and it’s going to be free because it’s going to run on the device you already bought,” he said. “It’s also going to be private, because you’re not going to send it to the cloud.”

That combination of cost savings and privacy could be a major shift for how people interact with AI daily. Instead of relying on cloud servers, a tiny LLM smartphone could handle reasoning tasks locally — no internet connection required.

PrismML’s progress from 95% to 98% benchmark parity in just a few months shows rapid improvement. If the company can maintain that trajectory with larger models, on-device AI could become far more capable than many expect.

For now, Bonsai 2 27B represents a meaningful step toward putting genuinely useful AI in your pocket — without the cloud, without the cost, and without sacrificing privacy.

Garry Tan Distillation: US Open-Weight AI Labs Should Distill
Nvidia Buys Hugging Face for $12.9 Billion
Google Pixel 11: Gemini AI and Camera Features
Claude CarPlay App: Anthropic AI Now in Your Car
Bee AI Note-Taking Devices: Eclipse and Frontier Debut

Sign Up for Our Daily Newsletter

Get the latest breaking news and updates delivered straight to your inbox.
By signing up, you agree to our Terms and Conditions and acknowledge the data practices in our Privacy Policy. You may unsubscribe at any time.
Share This Article
Facebook Copy Link Print
Share
Matilda Lucia, Techticia.com author avatar
ByMatilda
Follow:
At Techticia.com, I cover technology, gadgets, AI, product reviews and practical consumer tips, helping readers follow the tools and trends shaping the digital world.
Previous Article King Charles AI Summit: Urgent Warning to Control AI King Charles AI Summit: Urgent Warning to Control AI
Next Article Zoox Robotaxi Nevada: 100-Vehicle Cap to Expire Zoox Robotaxi Nevada: 100-Vehicle Cap to Expire
Leave a Comment

Leave a Reply Cancel reply

You must be logged in to post a comment.

Stay Connected

FacebookLike
XFollow
InstagramFollow
YoutubeSubscribe
WhatsAppFollow
LinkedInFollow
ThreadsFollow
BlueskyFollow

 banner							
 banner
Find your people. Find your next chapter.
Connect, explore jobs, and share what you love.
Get Semasocial

Latest News

Backglow MacBook App: Control the Glowing Apple Logo
Backglow MacBook App: Control the Glowing Apple Logo
Apple Gadgets
Sync Apple Music Across Devices: Easy Setup Guide
Sync Apple Music Across Devices: Easy Setup Guide
Apple Apps
Halide iPhone Duo Update Brings Foldable Camera Features
Halide iPhone Duo Update Brings Foldable Camera Features
Apple Gadgets
Qwen Model Upload: Cohere's Alibaba Test Explained
Qwen Model Upload: Cohere’s Alibaba Test Explained
AI

You Might also Like

Colossal Biosciences AI Platform Speeds Up De-Extinction

Matilda Lucia, Techticia.com author avatar
Matilda
6 Min Read
Smartphone displaying Claude AI app icon on screen
AI

Anthropic Claude Watermarks Anger Users: Hidden AI Text Code

Matilda Lucia, Techticia.com author avatar
Matilda
9 Min Read
Anthropic AI agents exploited websites: internet cut off
AI

Anthropic AI agents exploited websites: internet cut off

Matilda Lucia, Techticia.com author avatar
Matilda
7 Min Read
//

Craving the latest tech news? Techticia delivers! Dive deep into insightful articles on everything from cutting-edge gadgets and AI advancements to reviews, consumer tips, and future tech trends. Techticia.com keeps you informed and connected to the ever-evolving world of technology.

Sign Up for Our Newsletter

Subscribe to our newsletter to get our newest articles instantly!

Techticia.comTechticia.com
Follow US
© 2026 Techticia.com. All Rights Reserved.
Stay Ahead with Techticia
Get the latest technology news, reviews, guides, and updates from Techticia delivered to your inbox.
By subscribing, you agree to our Privacy Policy. Unsubscribe at any time.
Welcome Back!

Sign in to your account

Username or Email Address
Password

Lost your password?