Models & LabsInvestment · $22.25M
PrismML compresses LLMs to run locally with minimal loss
TechCrunch AIhigh confidence
Why it matters
- →Ternary weight compression achieves near-parity with uncompressed models, challenging the assumption that small models are weak.
- →A 5.9 GB footprint makes advanced reasoning models viable for consumer devices without cloud infrastructure.
- →Strong academic backing and early traction suggest this compression method could become a standard for local AI.
PrismML, a startup founded by Caltech researchers, has released Bonsai 2 27B, a highly compressed version of Alibaba's Qwen3.8 model that fits into 5.9 GB of memory. The company achieved this using ternary weight compression, reducing the model size by up to 10x while retaining 98% of the original benchmark performance. PrismML recently closed a $22.25 million seed round led by Khosla Ventures and Cerberus Capital to further develop its compression technology. The startup plans to apply this technique to larger models in the coming months, aiming for local deployment on PCs and smartphones.
Read originalThe story around this
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
How we got here
- 1JetBrains Releases Mellum2: Efficient 12B MoE Model
Hugging Face Blog · June 1, 2026 · Background
- 2Subquadratic claims breakthrough in LLM efficiency
MIT Technology Review AI · June 19, 2026 · Background
- 3ZML Launches Free Inference Software for AI Chips
TechCrunch AI · July 8, 2026 · Background
- 4PrismML Runs 27B-Parameter Model on iPhone
Lev Selector · July 17, 2026 · Related
- 5Together AI Introduces Autoscaling for LLM Inference
Together AI Blog · July 31, 2026 · Background
- 6Alibaba Challenges US AI with Qwen3.8-Max Model
The Verge AI · August 3, 2026 · Background
- 7Qwen3.8-27B Model Explored and Optimized
Sam Witteveen · August 18, 2026 · Related
- 8Granite 4.2 LLMs Introduced by IBM and Hugging Face
Hugging Face Blog · August 25, 2026 · Related
What happened next
- 9PrismML Tiny LLM Runs on Qualcomm Smart Glasses
TechCrunch AI · September 24, 2026 · Related
- 10Bonsai 2 and Needle: Tiny Local AI Models
Lev Selector · September 25, 2026 · Background
- 11Fine-tunes reduce Qwen3.8-27B reasoning tokens
Sam Witteveen · October 4, 2026 · Related
- 12llama.cpp adds Prism Bonsai 2 27B support
llama.cpp Releases · October 12, 2026 · Same story
- 13llama.cpp adds WebGPU Q2_0 support
llama.cpp Releases · October 22, 2026 · Background
More from TechCrunch AI
© TechCrunch AIMarket & Regulationother
Judge allows NAACP lawsuit over xAI pollution to proceed
xAI's attempt to dismiss the NAACP's Clean Air Act lawsuit failed after a federal judge ruled that its gas turbines are not exempt mobile sources and that plaintiffs have standing. This legal setback confirms that AI infrastructure operators cannot bypass environmental permitting by classifying heavy industrial equipment as temporary or mobile. The ruling forces xAI to face scrutiny over air quality impacts at its Colossus data center, setting a precedent for how large-scale compute facilities are regulated under existing environmental laws.
TechCrunch AI
© TechCrunch AIInvestment · $20B
Market & Regulationbusiness
OpenAI cuts funding target to $20B amid tepid demand
OpenAI is scaling back its ambitions as investor appetite for massive AI infrastructure bets cools. The company has lowered its fundraising target from $30 billion to $20 billion, with only $11 billion secured so far against a valuation that has dropped from $1.4 trillion. This retreat signals a growing wariness among Wall Street investors regarding the capital intensity of the AI race, forcing even the most dominant players to adjust their financial expectations.
TechCrunch AI
© TechCrunch AIMarket & Regulationbusiness
Nvidia weighs guaranteeing Nscale debt for Anthropic data center
Nvidia is considering a residual value guarantee for Nscale’s $44.6 billion Anthropic data center in West Virginia, effectively co-signing the construction debt to keep the project alive. This move signals deepening financial entanglement between chipmakers and neocloud providers as investors grow jittery about massive capital expenditures. With Nscale struggling to deploy chips and facing local opposition, Nvidia’s backing aims to secure a major customer for its next-gen Vera Rubin hardware while mitigating risk for lenders. It marks a significant escalation in Nvidia’s financial involvement with neocloud providers beyond standard chip purchases. The arrangement would see Nvidia step in if Nscale defaults on construction payments, helping secure financing for the facility that will house Anthropic’s infrastructure using Nvidia’s Vera Rubin chips. Nscale, which is seeking a $35 billion IPO, has faced delays and local opposition, while Anthropic recently filed for its own public offering amid concerns over future credit ratings. This potential guarantee underscores the industry's reliance on chipmaker balance sheets to sustain multi-billion dollar infrastructure buildouts.
TechCrunch AI
More in Models & Labs
Models & Labsmodels
llama.cpp b12029 adds CUDA 13 and Snapdragon support
This release quietly cements llama.cpp as the universal inference runtime by finally supporting NVIDIA's latest CUDA 13 stack alongside AMD's ROCm 10.0. The inclusion of native builds for Linux arm64 with Snapdragon chips, covering CPU, Adreno GPU, and Hexagon NPU, signals a serious push into mobile edge deployment. While Apple Silicon KleidiAI is disabled in this specific build, the broader hardware coverage means developers no longer need to compile from source for most modern accelerators. It’s less about new features and more about removing friction for anyone running local models on bleeding-edge or diverse hardware.
llama.cpp Releases
© The Verge AIModels & Labsother
Apple Intelligence arrives in China via iOS 27.3
Apple is finally unlocking its AI suite for Chinese users with the iOS 27.3 beta, marking a significant regulatory milestone after months of approval delays. The rollout relies on Alibaba’s Qwen model rather than OpenAI, integrating features like notification summaries and visual intelligence directly into the OS. By sidestepping the new Siri in favor of established text and image tools, Apple is prioritizing immediate utility over the full generative experience. This move effectively brings Apple’s AI capabilities to a massive new market, albeit with localized constraints that distinguish it from the global release.
The Verge AI
© Matt WolfeModels & Labsmodels
Google Releases Gemini 4 Carbon for Cost-Efficient Inference
Google launched Gemini 4 Carbon, a new model variant aimed at providing high performance at a lower cost for inference tasks.
Matt Wolfe