{"slug": "lfm2-5-2-6b-on-device-agents", "title": "LFM2.5-2.6B: On-Device Agents", "summary": "Liquid AI released LFM2.5-2.6B, a 2.6B dense model for on-device agentic workloads with a 128K context window and native tool calling, designed to run on edge devices and integrate with agent harnesses like Hermes Agent, OpenClaw, and Pi.", "body_md": "[← Back to Text Models](/lfm/models/text-models)LFM2.5-2.6B is Liquid AI’s 2.6B dense model built for agentic workloads with a 128K context window and native tool calling. It runs on edge devices and it is trained to work reliably inside agent harnesses like Hermes Agent, OpenClaw, and Pi.\n\n## Specifications\n\n## Agentic & Tool Use\n\nNative tool calling, trained inside real agent harnesses\n\n## 128K Context\n\nLong context for tool traces and multi-step workflows\n\n## On-device\n\nSmall enough to run on a laptop or phone\n\n## Quick Start\n\n- Transformers\n- llama.cpp\n- vLLM\n- SGLang", "url": "https://wpnews.pro/news/lfm2-5-2-6b-on-device-agents", "canonical_source": "https://docs.liquid.ai/lfm/models/lfm25-2.6b", "published_at": "2026-08-04 21:03:23+00:00", "updated_at": "2026-08-04 21:23:30.782705+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "ai-products"], "entities": ["Liquid AI", "LFM2.5-2.6B", "Hermes Agent", "OpenClaw", "Pi"], "alternates": {"html": "https://wpnews.pro/news/lfm2-5-2-6b-on-device-agents", "markdown": "https://wpnews.pro/news/lfm2-5-2-6b-on-device-agents.md", "text": "https://wpnews.pro/news/lfm2-5-2-6b-on-device-agents.txt", "jsonld": "https://wpnews.pro/news/lfm2-5-2-6b-on-device-agents.jsonld"}}