Liquid AI's smallest model yet LFM2.5-230M beats models 4X its size at data extraction, can run 'anywhere'

Liquid AI has released its smallest AI language model, LFM2.5-230M, a 230-million-parameter foundation model designed for on-device agentic workflows, which outperforms models 4X its size in data extraction and can run on devices such as smartphones, laptops, and robotics. The model utilizes the LFM2 architecture to achieve high inference speeds without massive memory overhead, making it suitable for edge devices. With a memory footprint of under 400MB, the model achieves decode speeds of 213 tokens per second on a Samsung Galaxy S25 Ultra and 42 tokens per second on a Raspberry Pi 5. This architectural efficiency has significant implications for engineers building AI systems, as it enables complex workflows on edge devices without requiring massive computational power or persistent cloud connections.

⚡ Key Takeaways

• LFM2.5-230M, a 230-million-parameter model, outperforms models like Alibaba Qwen3.5-0.8B and Google Gemma 3 1B in data extraction.
• The LFM2 architecture enables high inference speeds without massive memory overhead, making it suitable for edge devices.
• The model has a memory footprint of under 400MB and achieves decode speeds of 213 tokens per second on a Samsung Galaxy S25 Ultra and 42 tokens per second on a Raspberry Pi 5.
• The model supports an expansive 32K context window, allowing it to ingest substantial documents or continuous streams of robotic telemetry.
• The model operates under a dual-use commercial license, free for individuals and companies generating less than $10 million in annual revenue.

The release of LFM2.5-230M has significant implications for engineers building AI systems, as it enables complex workflows on edge devices without requiring massive computational power or persistent cloud connections. This architectural efficiency can lead to more efficient and cost-effective AI deployments, particularly in industries where data extraction and local processing are critical.

✅ Practical Steps

1. Evaluate the LFM2.5-230M model for use in data extraction and local deployment on edge devices.
2. Consider the LFM2 architecture for building lightweight data extraction pipelines and autonomous edge systems.
3. Assess the model's performance on specific devices, such as smartphones, laptops, and robotics, to determine its suitability for particular use cases.

Want the full story? Read the original article.