Arm
@Arm
Armโs foundational technology is defining the future of computing. A future built by the greatest technology ecosystem in the world. A future built on Arm.
๊ฐ์
April 2009
1.8K ํ๋ก์ ์ค 101.3K ํฌ
๐Congratulations to @liquidai on the launch of LFM3, its new family of compact, open-weight models spanning text, vision, audio and decision-making!
We're proud to support LFM3 from day one, building on our collaboration with Liquid AI to advance efficient, on-device AI inference on Arm.
Early, provisional benchmarks of LFM3-1.7B (Q4_0) on Arm CPUs show a 3.1ร speedup in prefill and 1.5ร speedup in decode with SME2 enabled, achieving over 400 tokens/s prefill and over 50 tokens/s decode using just two CPU threads.
๋ ๋ณด๊ธฐ
Introducing ๐ป๐ต๐ผ๐น: our new generation of Liquid Foundation Models. With a new tri-hybrid architecture and new multimodal encoders, the LFM3 family is the ideal model suite for specialized intelligence.
> 6 model instances, across text, audio, vision, and omni modalities, generative and decision models.
> ๐ป๐ต๐ผ๐น-๐ท.๐ฝ๐ฑ: text, 256k context window.
> ๐ป๐ต๐ผ๐น-๐
๐ป-๐ธ๐ฑ: vision-language model with video-input support for spatial understanding
> ๐ป๐ต๐ผ๐น-๐ฐ๐๐๐๐-๐ธ๐ฑ: tool calls, voice chat, reasoning in 18 languages.
> ๐ป๐ต๐ผ๐น-๐พ๐๐๐-๐ธ๐ฑ: an omni model combines vision, audio, and language.
> ๐๐ท-๐ธ๐ฑ and ๐๐ท-๐พ๐๐๐-๐ธ๐ฑ: decision models with typed outputs and calibrated probabilities in a single pass. Omni adds audio input.
> fine-tune on 1 GPU.
> a new tri-hybrid architecture
> architecture optimized for 14 processor classes across GPUs, CPUs, and NPUs.
> ๐ป๐ต๐ผ๐น-๐
๐ป-๐ธ๐ฑ is a beast. It performs competitively with Gemma 4 12B (6x larger), and Gemini 3.5 Flash Lite, and GPT-5.6 Luna, with 1.9B parameters.
> ๐ป๐ต๐ผ๐น-๐ท.๐ฝ๐ฑ is #1# on the Pareto frontier of device intelligence ๐๐๐๐๐๐๐.๐๐๐๐๐๐.๐๐
Every modality for specialized intelligence. Pick the one that fits your use case, customize it, and run it on-device.
1/ ๐งต
๋ ๋ณด๊ธฐ
