Models & Labs
Liquid AI releases LFM3 multimodal models
Hugging Face Bloghigh confidence
Why it matters
- →LFM3 proves that specialized 2B models can outperform generalist 9B+ models in specific domains like agentic tasks.
- →Custom hardware-aware encoders drastically reduce the compute cost for vision and audio processing on edge devices.
- →The architecture enables complex reasoning and video understanding to run locally, reducing reliance on cloud inference.
Liquid AI has released LFM3, a new generation of Liquid Foundation Models designed for specialized, on-device intelligence. The family includes text, vision, audio, omni, and decision models, all branching from a 1.7B parameter base. Key technical improvements include a tri-hybrid architecture and custom encoders that reduce compute requirements significantly; the vision model uses 3x less compute than previous standards while matching larger models' accuracy. The text model demonstrates superior agentic performance, outperforming Qwen3.5-9B on WebArena benchmarks despite having far fewer parameters.
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
- 1vLLM v0.25.0 Release Enhances Model Execution
vLLM Releases · July 13, 2026 · Related
- 2Together AI launches Inkling model on day 0
Together AI Blog · July 15, 2026 · Related
- 3Thinking Machines releases Inkling model
Fireship · July 20, 2026 · Background
- 4Startups Innovate Beyond Transformers in LLMs
MIT Technology Review AI · August 10, 2026 · Related
- 5Together AI Enables A/B Testing for LLMs
Together AI Blog · August 17, 2026 · Related
- 6GLM 5.3 Flash: Z.AI's Cost-Efficient Model Variant
Sam Witteveen · August 30, 2026 · Related
- 7Liquid AI opens d1 decision models for edge inference
Hugging Face Blog · October 7, 2026 · Same story
- 8Liquid AI releases efficient on-device agent model
Hugging Face Blog · October 13, 2026 · Same story
What happened next
- 9llama.cpp adds LiquidAI LFM3-Audio and Omni support
llama.cpp Releases · October 28, 2026 · Related
Follow this story
Open the full story →Liquid AI releases LFM3 multimodal models
3 developments
- Oct 27 · Hugging Face Blog
Liquid AI releases LFM3 multimodal models (This article) - Oct 28 · llama.cpp Releases
llama.cpp adds LiquidAI LFM3 model support↳ llama.cpp adds support for LiquidAI LFM3 models to enable local inference on Apple Silicon and AMD ROCm GPUs. - Oct 30 · llama.cpp Releases
llama.cpp adds OCP FP8 E4M3 support↳ llama.cpp adds OCP FP8 E4M3 quantization support with fast CPU kernels and fixes for fused QKV projections.
More from Hugging Face Blog
© Hugging Face BlogAgentsagents
NVIDIA Nemotron powers interactive Halloween agent
An NVIDIA engineer built a fully local voice agent for a Halloween attraction using the new Nemotron Voice Agent framework. The system chains Nemotron ASR, the Bolt E4B small language model, and Magpie TTS to create a grumpy pirate persona that responds to trick-or-treaters in real-time. Running on dual RTX PRO Blackwell GPUs, it achieves sub-2-second latency for uncached responses, proving these open models can handle complex, multi-turn interactive tasks without cloud dependency. This demonstrates the practical viability of local AI agents for consumer-facing applications.
Hugging Face Blog
© Hugging Face BlogModels & Labsmodels
Granite Switch 4.1.1 patches adapters without retraining base
IBM’s Granite Switch 4.1.1 proves modular architectures can move with the speed of standard software updates. By keeping the base model untouched and only updating specialized adapters like the multilingual ASR and decision logic, they sidestep the massive compute cost of full retraining. This approach lets developers compose new expertise on a single endpoint without migration headaches. It signals a shift toward composable AI where models evolve incrementally rather than in disruptive leaps.
Hugging Face Blog
© Hugging Face BlogGeneral AIresearch
Project ProtoStar launches to make public science AI-ready
Hugging Face and a coalition of major universities are launching Project ProtoStar to solve the 'last mile' problem of scientific data: making it machine-actionable. With $190 billion in annual U.S. federal R&D funding, the initiative pushes institutions like CMU and Purdue to release datasets in standardized formats rather than static files. The kickoff challenge tests AI's ability to predict urban climate conditions using historical weather records from 210 cities. This shifts the focus from mere publication to actual interoperability, creating a reusable infrastructure for AI-driven discovery.
Hugging Face Blog
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