A new family of 3-valued LLMs, "Ternary Bonsai 2," has been released in 2B, 4B, and 9B parameter sizes. These models are capable of processing up to 120 tokens per second on an iPhone. This development represents a step towards more efficient and accessible large language models. AI
IMPACT Offers potential for faster and more efficient on-device AI processing.
RANK_REASON Release of a new LLM family with specific performance metrics. [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 →