A 2.6B model with tool calling and 128K context now runs at 30 tok/s on a phone
英文标题：A 2.6B model with tool calling and 128K context now runs at 30 tok/s on a phone
Liquid AI released LFM2.5-2.6B today, and this might be more relevant to local AI than another massive model most people cannot run. The model is only 2.69B parameters, has 128K context, supports tool calling and was post-trained specifically for multi-step agent workflows. The official Q4_K_M
打开原文相关 5 · 机会 3
中文
Liquid AI released LFM2.5-2.6B today, and this might be more relevant to local AI than another massive model most people cannot run. The model is only 2.69B parameters, has 128K context, supports tool calling and was post-trained specifically for multi-step agent workflows. The official Q4_K_M
中文全文翻译正在补齐，可先查看下方英文原文。
English / 原文
Liquid AI released LFM2.5-2.6B today, and this might be more relevant to local AI than another massive model most people cannot run. The model is only 2.69B parameters, has 128K context, supports tool calling and was post-trained specifically for multi-step agent workflows. The official Q4_K_M