PulseAugur
中
实时 14:31:16
English(EN) LFM2.5-24B-A2B: Built for Fast On-Device Agents https://www. liquid.ai/blog/lfm2-5-24b-a2b # ai

Liquid AI 发布 LFM2.5-24B-A2B,用于快速设备端智能体

Liquid AI 发布了 LFM2.5-24B-A2B,这是一款拥有 240 亿参数的专家混合(MoE)模型,专为设备端智能体设计。该模型每个 token 激活约 20 亿参数,并针对多步工具使用进行了优化,而非通用聊天。它在智能体和工具使用基准测试中表现强劲,在 AutomationBench 等特定任务上优于更大的模型,而 Qwen3.6-35B-A3B 等更大的模型在知识和编码任务上表现更好。LFM2.5-24B-A2B 能够运行在 Apple M5 Max 和 AMD Ryzen AI Max+ 处理器等消费级硬件上,提供快速的本地工具调用和指令遵循能力。 AI

影响 使更强大的 AI 智能体能够在消费级硬件上高效运行,可能加速设备端 AI 应用。

排序理由 前沿实验室模型发布,附带系统卡。

在 Mastodon — mastodon.social 阅读 →

AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →

Liquid AI 发布 LFM2.5-24B-A2B,用于快速设备端智能体

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Frontier Release
前沿实验室模型发布,附带系统卡。
Source corroboration
4 independent sources
Strong cross-source corroboration — multiple independent publishers covered this within the clustering window.
Topics
model release, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
8 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.
Coverage growth since scoring
+1 source(s) since last score
New sources have picked up this story since our last re-score. Score will update on the next scoring pass.

完整方法见我们的编辑标准。

报道来源 [4]

  1. Hacker News — AI stories ≥50 points TIER_1 English(EN) · gmays ·

    LFM2.5-24B-A2B:专为快速设备端智能体而构建

  2. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Liquid AI 发布 LFM2.5-24B-A2B:面向设备端代理的 2B 激活 MoE

    <p>Liquid AI released LFM2.5-24B-A2B, an open-weight 24B Mixture-of-Experts reasoning model with 2B active parameters. It targets local agents, scoring 48.09 on LiveClaw and decoding 184 tokens/s on an Apple M5 Max.</p> <p>The post <a href="https://www.marktechpost.com/2026/10/14…

  3. Mastodon — mastodon.social TIER_1 English(EN) · h4ckernews ·

    LFM2.5-24B-A2B:专为快速设备端智能体打造 https://www.liquid.ai/blog/lfm2-5-24b-a2b 评论:https://news.ycombinator.com/item?id=50097369 #HackerNews

    LFM2.5-24B-A2B: Built for Fast On-Device Agents https://www. liquid.ai/blog/lfm2-5-24b-a2b Comments: https:// news.ycombinator.com/item?id=5 0097369 # HackerNews # LFM2 .5 # 24B # OnDevice # AI # FastAgents # LiquidAI

  4. Mastodon — mastodon.social TIER_1 English(EN) · CuratedHackerNews ·

    LFM2.5-24B-A2B:专为快速设备端智能体而构建 https://www. liquid.ai/blog/lfm2-5-24b-a2b # ai

    LFM2.5-24B-A2B: Built for Fast On-Device Agents https://www. liquid.ai/blog/lfm2-5-24b-a2b # ai