Tencent has launched Hy3, a new open-source language model. This model boasts 295 billion parameters but utilizes a mixture-of-experts architecture, meaning only 21 billion parameters are active at any given time. Tencent claims Hy3 performs comparably to models five times its active size and has halved its hallucination rate to 5.4%. AI
IMPACT This model's efficient parameter usage could influence future open-source LLM development and deployment strategies.
RANK_REASON Release of an open-source model with specific technical details. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
