AI Briefing#liquid-ai
The latest AI and developer news about #liquid-ai, with the original source and a short summary.
The latest AI and developer news about #liquid-ai, with the original source and a short summary.
The 74M parameter codec uses Grouped FSQ to enable parallel prediction, outperforming Mimi and NanoCodec in reconstruction quality and speech generation metrics.
Liquid AI has released six 2B-scale LFM3 models supporting execution on edge devices and multimodal processing.
Runs on laptops with 2.45B active parameters, outperforming Qwen3.6 in tool-calling performance
The multimodal model returns structured probabilities for classification and routing tasks rather than generating text.
The open-weight model achieves 184 tokens/s on an Apple M5 Max and scores 48.09 on LiveClaw, outperforming larger models in agentic tasks.
The new open-weight model features only 2B active parameters, achieving up to 1.7x faster inference than comparable MoEs on-device while matching larger models on key agentic benchmarks.
The d1 model returns structured answers with probabilities for classification and routing without generating text tokens.
The new d1-3B model achieves a score of 48.57 on the Decision Index 0.2.1, outperforming larger models like Decider 35B-A3B.
A Jev-compatible API-based model offering low latency and calibrated probabilities with 0 output tokens
Liquid AI has released LFM2.5-VL-3B-DSpark, a draft model for speculative decoding of its VLM LFM2.5-VL-3B, which improves decoding speed by up to 3.13x on the Apple M5 Max.
The new 280M-parameter draft model accelerates LFM2.5-VL-3B inference by up to 3.13x on device and 2.66x on H100 GPUs.
Artificial Analysis, in collaboration with Liquid AI, released a benchmark measuring the intelligence and inference speed of small LLMs under 8GB on the iPhone 17 Pro.
Liquid AI has released the LFM2.5 Q4_0 models, which minimize quality degradation through Quantization-Aware Distillation (QAD).
Liquid AI conducted an experiment to verify the autonomous production problem-solving capabilities of coding agents.
Liquid AI released a 2.6B agent model capable of running on mobile phones and CPUs.
Liquid AI has released two LFM2.5 encoder models capable of fast long-context processing even in CPU environments.
Liquid AI has unveiled a technique that expands the vocabulary of the LFM2.5-8B-A1B model from 65K to 128K without retraining from scratch.
Liquid AI has launched LFM 2.5-230M, an ultra-lightweight parameter model optimized for agentic workflows.
Liquid AI has released highly efficient multilingual embedding and ColBERT models to improve RAG performance.
Liquid AI has released an on-device 8B-A1B MoE model trained on 38T tokens with enhanced reasoning capabilities.
Liquid AI has released LFM2.5-8B-A1B, an edge model with a 128K context and 38T tokens of training data.
LiquidAI has released LFM2.5-8B-A1B, a hybrid model designed for on-device deployment.