阿布達比 TII 發表 Falcon 2 11B：搭載 5 兆 Token 訓練的預訓練語言與視覺語言模型
Original: Falcon 2: An 11B parameter pretrained language model and VLM, trained on over 5000B tokens and 11 languages
The Technology Innovation Institute (TII) of Abu Dhabi has officially released a new open-source model family on Hugging Face — Falcon 2…
阿布達比技術創新研究所（TII）推出全新開源模型 Falcon 2 11B，包含預訓練語言模型與視覺語言模型（VLM）。該模型在超過 5 兆（5000B）Token 的高質量數據上進行訓練，支援 11 種語言。其性能在同量級模型中表現亮眼，並採用 Apache 2.0 授權，對開源社群與開發者極具吸引力。
The Technology Innovation Institute (TII) of Abu Dhabi has officially released a new open-source model family on Hugging Face — Falcon 2 11B. This model, with 11 billion parameters, represents the latest evolution of the Falcon series and aims to set a new performance benchmark for mid-sized models.
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.