ainewsblitz.com

Breaking

Liquid AI Releases LFM2.5-24B-A2B, a Sparse 24B Model That Outperforms Larger Competitors on Agent Tasks

  • Foundation Models
  • AI Agents
  • Open Source

Liquid AI has released LFM2.5-24B-A2B, a 24-billion-parameter mixture-of-experts model with only 2.45 billion parameters active per token, positioned to challenge larger competitors in on-device agentic workloads. Published on October 13, 2026, as open-weight checkpoints on Hugging Face, the model represents the largest member of Liquid's LFM2.5 family and signals a strategic shift toward reasoning-focused, locally-deployable agents.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 26,525 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year