Liquid AI, an MIT spinout valued at roughly $2 billion, released its LFM2.5-230M foundation model on June 25, targeting on-device AI workflows. The model, with 230 million parameters, scored 22.51 on data extraction tasks using the CaseReportBench dataset, outperforming Alibaba’s Qwen3.5-0.8B (13.83) and Google’s Gemma 3 1B (2.28). Pre-trained on 19 trillion tokens, it runs at 213 tokens per second on a Samsung Galaxy S25 Ultra and 42 tokens per second on a Raspberry Pi 5. The pre-trained weights are available on Hugging Face. Liquid AI was founded in December 2023 by Ramin Hasani, Mathias Lechner, Alexander Amini, and Daniela Rus. The company raised $37.5 million in seed funding and closed a $250 million Series A led by AMD Ventures by December 2024, pushing total funding to approximately $293 million. The company has no direct blockchain connections.