StarCoder2 與 The Stack v2 正式發布：新一代開源程式碼大模型與超大資料集
Original: StarCoder2 and The Stack v2
The BigCode community, jointly led by Hugging Face and ServiceNow, together with NVIDIA, has officially announced the launch of a new…
由 Hugging Face 與 ServiceNow 主導的 BigCode 社群，聯合 NVIDIA 正式發布了新一代開源程式碼大語言模型 StarCoder2。該系列提供 3B、7B 和 15B 三種參數版本，並同步推出全新升級的 The Stack v2 資料集。StarCoder2 支援超過 600 種程式語言，並在程式碼生成、補全與修復等任務上展現出同量級模型中的頂尖性能，持續推動負責任且開放的 AI 開發。
The BigCode community, jointly led by Hugging Face and ServiceNow, together with NVIDIA, has officially announced the launch of a new generation of open-source code large language model series called "StarCoder2," along with the brand-new, massive-scale dataset used to train it: "The Stack v2."
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.