Google 推出全新開源大語言模型 Gemma 2：9B 與 27B 雙版本，效能超越同級模型
Original: Welcome Gemma 2 - Google’s new open LLM
Google has officially launched the next generation of its open-source large language model, Gemma 2, with an initial release in two sizes —…
Google 聯合 Hugging Face 推出新一代開源模型 Gemma 2，提供 9B 與 27B 兩種參數規格。27B 版本在同量級中表現極為優異，甚至能與體積兩倍大的模型競爭；9B 版本也超越了 Llama 3 8B。此系列模型採用知識蒸餾與混合注意力機制，並已在 Hugging Face 生態系中獲得全面支援。
Google has officially launched the next generation of its open-source large language model, Gemma 2, with an initial release in two sizes — 9B (9 billion parameters) and 27B (27 billion parameters) — aiming to give developers and researchers a choice that combines high performance with deployment flexibility.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.