# Liquid AI — LFM2.5-2.6B: Deploy Agents Everywhere

- Company: Liquid AI (liquid.ai)
- Announced: 2026-08-04
- Category: new-model
- Coverage: not counted
- Announcement: yes
- Group: models
- Source: https://www.liquid.ai/blog/lfm2-5-2-6b
- Record: https://forck.live/items/16776-lfm2-5-2-6b-deploy-agents-everywhere
- Subject: LFM / d1 models
- Open weights: yes
- Models affected: LFM2.5-2.6B, LFM2.5-2.6B-Base
- Context window: 128K context-extension phase

Liquid AI released LFM2.5-2.6B, a 2.6 billion parameter model designed for on-device agentic workloads such as planning, tool calling, and multi-step tasks. The model is pre-trained on approximately 34 trillion tokens, supports a 128K token context window, and is available on Hugging Face in both base and post-trained versions. Benchmarks show it is competitive with models nearly four times its size on STEM, instruction following, tool use, and agentic tasks.

## Evidence

Verbatim from https://www.liquid.ai/blog/lfm2-5-2-6b:

> Today, we release LFM2.5-2.6B , an agentic model that runs entirely on-device. It is small enough to run on a phone, fast enough to stay responsive on a CPU, and capable enough to power agentic workflows: planning, calling tools, and tackling multi-step tasks.

---

Record: https://forck.live/items/16776-lfm2-5-2-6b-deploy-agents-everywhere
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md