ainewsblitz.com

Breaking

Nanbeige4.2-3B, a 3B Agent Model Trained on 28 Trillion Tokens, Outperforms Rivals Three to Four Times Larger

Nanbeige4.2-3B, a 3B Agent Model Trained on 28 Trillion Tokens, Outperforms Rivals Three to Four Times Larger
  • Foundation Models
  • AI Agents
  • Open Source

Nanbeige LLM Lab, a research unit of BOSS Zhipin, has released Nanbeige4.2-3B, a compact agentic language model that its developers say beats far larger systems such as Qwen3.5-9B and Gemma4-12B on a range of agent benchmarks. The model landed simultaneously on Hugging Face and ModelScope on July 22, 2026, under an Apache-2.0 license, positioning it as an open, locally runnable option for tool use, coding, reasoning, office work and deep research.

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 27,934 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year