No cloud, no GPUs, no problem: Liquid AI's new model LFM2.5-2.6B brings powerful AI agents to devices as small as a Raspberry Pi

Liquid AI debuted LFM2.5-2.6B, a new open-weight language model designed for agentic workloads, earlier this week. The company said the model can run entirely on local hardware, from smartphones and laptops down to a Raspberry Pi, without relying on cloud inference or GPUs. Maxime Labonne, Liquid AI's head of post-training, said in an interview with VentureBeat that the model runs very well on CPUs and works pretty fast on a Raspberry Pi. The model contains 2.6 billion parameters, supports a 128,000-token context window, and includes native tool calling. Both the post-trained model and a base checkpoint are available now on Hugging Face with day-one support for major inference stacks including llama.cpp, MLX, vLLM, SGLang, and ONNX.