Tech article
ZAYA1-8B: An 8B Moe Model with 760M Active Params Matching DeepSeek-R1 on Math
Latest tech news from Hacker News on NeuralNews: ZAYA1-8B: An 8B Moe Model with 760M Active Params Matching DeepSeek-R1 on Math.
Hacker News | May 7, 2026 | steveharing1