The transformers library has released version 5.17.0 Hy4-Preview, featuring a 780 billion parameter mixture-of-experts language model. This advanced model activates 49 billion parameters for each inference, indicating a significant step in model efficiency and capability. AI
IMPACT This release introduces a large mixture-of-experts model, potentially offering new capabilities for AI development and research.
RANK_REASON Release of a new model version and its technical specifications. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →