答案已發布昨天Last edited 昨天16 來源

Google NotebookLM 終於識自己做嘢：升級 Gemini 3.5 後識寫 Code、自己搵 Source 同整 Excel

Google 嘅 2026 年 6 月 8 日更新全面升級 NotebookLM，用咗 Gemini 3.5 同 Antigravity，俾佢有自主代碼執行、傾偈搵料等 Agent 功能。新 NotebookLM 識得自己上網搵 Source，唔使吓吓自己 Upload 文件，解決咗開頭唔知點搵料嘅煩惱。

使用 Studio Global AI 搜尋並查核事實瀏覽更多熱門頁面

78K0

What are the key features, capabilities, and availability details of the major NotebookLM upgrade Google announced on June 8, 2026, which inAI-generated editorial hero image for What are the key features, capabilities, and availability details of the major NotebookLM upgrade Google announced on June 8, 2026, which in.
AI 提示
Create a landscape editorial hero image for this Studio Global article: What are the key features, capabilities, and availability details of the major NotebookLM upgrade Google announced on June 8, 2026, which in. Article summary: Based on the available sources, here's a comprehensive breakdown of the June 8, 2026 NotebookLM upgrade. [6]. Topic tags: general, general web, user generated. Reference image context from search candidates: Reference image 1: visual subject "A graphic depicting several overlapping dark-themed app screens on a green circuit board background, demonstrating how to convert study notes into videos and podcasts using a tool" source context "NotebookLM Update 2026: Every New Feature Explained" Reference image 2: visual subject "Google NotebookLM has had one of the most packed update cycles of any AI tool in recent memory. Since October 2025, the platform has shipped
openai.com

Google 呢次推出嘅更新，可以話係 NotebookLM 自推出以嚟最大規模嘅一次，直接將佢由一個乖乖坐喺度等人餵資料嘅研究助手，變成一個會主動出擊嘅自主研究 Agent。呢個喺 2026 年 6 月 8 號公布嘅升級，主要靠兩樣嘢撐起：新嘅預設模型 Gemini 3.5，同埋 Google 專為 Agent 而設嘅開發環境 Antigravity。

結果就係一個而家識得獨立執行代碼嚟分析你嘅數據、會主動上網搵 Source 畀你個新 Project，仲識將佢搵到嘅嘢包裝成靚靚嘅專業檔案格式嘅工具——而呢一切，都係喺每個 Notebook 入面嘅安全隔離雲端電腦入面搞掂。

不過要注意，呢次更新雖然係全球即時推出，但目標用家好有針對性：淨係畀 Google AI Ultra 訂閱戶同埋 Google Workspace 企業及商業客戶用。免費版用家幾時有得用，Google 暫時隻字不提。

換咗個新腦：Gemini 3.5 同 Antigravity 嘅化學作用

最大嘅改變係 NotebookLM 嘅基本推理引擎轉咗做 Gemini 3.5。 Google 話呢個係「準確性同可靠性」嘅一大步，仲話會「更加睇得清楚佢嘅思考過程」。早期嘅跑分結果都證明咗呢個講法。喺幾個重要測試入面，新版 NotebookLM 平均「勝率」超過 65%，比舊版高出大約 15 個百分點。喺專門啲嘅任務度，表現仲癲：深入文件分析勝率有 69.9%，而高階網頁研究同搵 Source 嘅勝率更加去到 78.2%，係一個好誇張嘅數字。

Studio Global AI

Search, cite, and publish your own answer

Use this topic as a starting point for a fresh source-backed answer, then compare citations before you share it.

使用 Studio Global AI 搜尋並查核事實

人們還問