Liquid AI launches LFM3: six-model family with first fully omni-modal model
avapamini · x · 2026-10-28
Liquid AI announced LFM3, a new generation of Liquid Foundation Models built on a tri-hybrid architecture with new multimodal encoders, emphasizing maximized intelligence per watt. The family spans six instances:
- LFM3-1.7B: text, 256k context window
- LFM3-VL-2B: vision-language with video input for spatial understanding
- LFM3-Audio-2B: tool calls, voice chat, reasoning in 18 languages
- LFM3-Omni-2B: the company's first fully omni model combining vision, audio, and language
- d1-2B and d1-Omni-2B: decision models outputting typed decisions with calibrated probabilities in a single pass (Omni adds audio input)
Positioned as a model suite for specialized, likely on-device intelligence.
Related event: LiquidAI Releases LFM3 Family of On-Device Models(15 posts)→
More from Models
- ChatGPT Work sync bug makes messages vanish from view while the model still recalls them — LynxAirSound · 2026-10-28
- Palisade Research finds Alibaba's Qwen 3.8 models routed some requests to company servers — The Decoder · 2026-10-28
- DeepSeek on $75K of hardware: 2,120 tok/s decode at 16 concurrent sessions — victormustar · 2026-10-28
- Frontier models hit 73.3% on SpatialBench 10 months after release, near 75% ceiling — kenbwork · 2026-10-28
- 5 models tested as Salesforce agents: Opus 87%, DeepSeek 82% at 1/75 the cost — TruthDeep8659 · 2026-10-28
- LiquidAI Ships 2B Omni Open Model With Day-One llama.cpp Support — ilintar · 2026-10-28