Zyphra's ZAYA1-8B model matches top AI benchmarks with under 1B parameters

Zyphra has released ZAYA1-8B, an open-source model that achieves performance comparable to DeepSeek-R1 on math benchmarks. The model also demonstrates competitive reasoning capabilities against Claude Sonnet 4.5 and approaches Gemini 2.5 Pro in coding tasks. Notably, ZAYA1-8B operates with fewer than one billion active parameters. AI

影响 Offers a new, efficient open-source model for math and coding tasks, potentially lowering barriers for researchers and developers.

排序理由 Open-source model release from a non-frontier lab with benchmark results. [lever_c_demoted from research: ic=1 ai=1.0]

在 Mastodon — sigmoid.social 阅读 →

• Claude Sonnet 4.5
• DeepSeek-R1
• Gemini 2.5 Pro
• ZAYA1-8B
• Zyphra

• 模型发布
• 论文

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →