Liquid AI has launched LFM2.5-230M, their most compact model yet. This open-weight model, with 230 million parameters, is designed for on-device operation on edge hardware. It achieves speeds of 213 tokens/s on a Samsung Galaxy S25 Ultra and 42 tokens/s on a Raspberry Pi 5, supporting tool use and data extraction. AI
IMPACT Enables more capable on-device AI applications, particularly for tool use and data extraction on consumer hardware.
RANK_REASON Release of a small, specialized model for on-device inference.
Read on Mastodon — fosstodon.org →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
