歡迎 Gemma：Google 全新開源大語言模型登陸 Hugging Face
Original: Welcome Gemma - Google’s new open LLM
Google has officially released a new family of open-source large language models called "Gemma" — a series of lightweight, state-of-the-art…
Google 正式推出全新開源模型家族 Gemma，包含 2B 與 7B 兩種參數大小，並提供預訓練與指令微調版本。 Gemma 採用與 Gemini 相同的技術與架構，在多項基準測試中表現優於同尺寸的 Llama 2 與 Mistral。 Hugging Face 已同步支援 Gemma，開發者可直接透過 Transformers、TGI、PEFT 等工具進行部署與微調。
Google has officially released a new family of open-source large language models called "Gemma" — a series of lightweight, state-of-the-art open-source models built using the same technology and research that underpins the Gemini models. Gemma is offered in two parameter sizes: 2B (2 billion parameters) and 7B (7 billion parameters). Each size includes both a base pre-trained (Base) and an instruction-tuned (IT) version, and is released under a permissive license that allows commercial use.
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.