English
This page is also available in

Home › Search and research › LFM2.5-24B-A2B

Model

LFM2.5-24B-A2B

LiquidAI/LFM2.5-24B-A2B

Search and research Source: Hugging Face modelsScore 1.1trending 440 (455 likes)· First listed Oct 15, 2026

Visit the LFM2.5-24B-A2B website / project page →

What is LFM2.5-24B-A2B

LFM2.5-24B-A2B released by LiquidAI is a 24B-parameter open-source large language model based on Liquid Foundation Models, supporting efficient text generation and inference tasks.

Key features

  • 24B-parameter model that activates only 2B parameters, supporting efficient text generation and inference
  • Designed for fast, private Agents on device, excelling at agentic and tool-use tasks
  • Always thinks before answering during inference, using only 2B active parameters per token
  • Supports a context length of up to 131072 tokens
  • Supports multiple languages including English, Arabic, Chinese, French, German, Japanese and Korean

Who it is for

Developers who need to run fast, private Agents on device, and application scenarios requiring efficient inference and tool-use capabilities.

How to get started

  1. Visit https://huggingface.co/LiquidAI/LFM2.5-24B-A2B to download the model weights
  2. Load the LFM2.5-24B-A2B model with Hugging Face Transformers for inference

Compiled from public descriptions on the tool's official website / README / launch page. Pricing and user numbers have not been verified; please check the official website. Some sections were drafted with AI assistance. Written in Chinese and English; other languages are machine-translated.

Similar alternatives (Search and research)

blumify 3.82
Blumify is a tool that tracks statements and shifts in opinion from AI leaders, allowing you to see how their positions on various AI topi…
DeepSeek-V4.1-Flash-1017 3.25
DeepSeek-V4.1-Flash-1017 is an open-source large language model released by DeepSeek, supporting efficient inference and multilingual capa…
Parsimony 3.21
Parsimony is an open-source benchmark that measures how much additional code AI coding models introduce when completing tasks, evaluating…
DishwasherBench 2.03
DishwasherBench is a language model benchmark game based on realistic dishwasher physics simulation. It can be played directly in the brow…
Hy-Memory-30B-A3B 0.93
Hy-Memory-30B-A3B released by Tencent is a Mixture-of-Experts (MoE) large language model available on Hugging Face.
absent-author 0.75
Absent Author is an open-source project that provides an evidence checklist and Agent skills for detecting AI-generated papers and determi…
MiniCPM-V-4.7-35B-A3B 0.69
MiniCPM-V-4.7-35B-A3B is an open-source multimodal large model released by OpenBMB. It supports multimodal understanding of images, video…
open-academic-paper-gen 0.68
A multi-agent pipeline tool that can generate academic paper drafts grounded in real literature based on a research topic, with citation h…

Get a “Featured on AI Hot List” badge · Submit a tool

← Back to “Search and research” · Full hot list · Use-case lists