News and Updates

NOV 5, 2026External article

Liquid AI now in Production in Mercedes Cars

11.05.2026
External articleLiquid AI now in Production in Mercedes Cars
10.27.2026
ModelsIntroducing LFM3: Every Modality for Specialized Intelligence
10.27.2026
PublicationsPerceiving and Predicting Dynamic Sports: Spatiotemporal Intelligence in Vision Models for Egocentric Tennis Video
10.22.2026
NewsRIPL: Real-time Intelligence When Every Millisecond Counts
10.15.2026
Modelsd1-ultra: Introducing the World’s Best Decision Model
10.13.2026
ModelsLFM2.5-24B-A2B: Built for Fast On-Device Agents
10.07.2026
ModelsOpen d1: Edge decision models for text, vision, and audio
10.05.2026
ModelsIntroducing d1: The most capable decision model, now with vision
09.29.2026
PublicationsLooped Actor: Depth-Recurrent Reasoning Models for Reinforcement Learning
09.24.2026
ModelsLFM2.5-VL-DSpark: Accelerating vision-language models on edge and beyond
09.23.2026
NewsLiquid AI, in Collaboration with Qualcomm Technologies, Brings Personal AI Context to Devices Powered by Snapdragon
09.17.2026
EngineeringSetting the Frontier of Aging Biology with Liquid Foundation Models
09.17.2026
PublicationsAn open benchmark and language models for AI in aging biology
08.24.2026
NewsIntroducing Pipette: A benchmarking suite for on-device intelligence
08.20.2026
ModelsLFM2.5-DSpark: Up to 3.2x Faster Inference from H100 to MacBook
08.19.2026
ModelsLFM2.5 Q4_0: Quantization-Aware Distillation for Edge Deployment
08.18.2026
EngineeringDesigning Loops for Production-Grade Work
08.14.2026
PublicationsBatch-wise Adaptive Pruning: Periodic Neuron Activation-Aware Weight Pruning for Language Reasoning Model
08.12.2026
ModelsLFM2.5-VL-3B: A Better and Faster Vision-Language Model for the Edge
08.11.2026
PublicationsMMAI Gym for Science: Training Liquid Foundation Models for Drug Discovery
08.05.2026
NewsMacPaw Partners with Liquid AI to Bring On-Device AI to Millions of Mac Users
08.04.2026
ModelsLFM2.5-2.6B: Deploy Agents Everywhere
07.28.2026
ModelsLFM2.5-Encoders: Fast at Long Context, Even on CPU
07.21.2026
EngineeringTokenizer Expansion: Upgrading a Model's Tokenizer in Place
07.16.2026
PublicationsIn-Place Tokenizer Expansion for Pre-trained LLMs
07.08.2026
PublicationsMulti-Agent Robotic Control with Onboard Vision-Language Models
07.07.2026
EngineeringReducing Doom Loops with Final Token Preference Optimization
06.30.2026
EngineeringIFStruct: Measuring structured-output compliance
06.25.2026
ModelsLFM2.5-230M: Built to Run Anywhere
06.18.2026
ModelsLFM2.5 Retrievers: Bi-directional LFMs for Fast Multilingual Search
06.17.2026
PublicationsDango: Predicting higher-order genetic interactions