llama
US /ˈlɑmə/
・UK /'lɑ:mə/
C1 高級英檢高級
n.名詞羊駝；美洲駝
We saw a llama during our Peruvian visit
影片字幕
黃仁勳暢談資料處理與生成式AI的未來！ (黃仁勳談數據處理與生成式AI的未來)
- LLAMA 2, now LLAMA 3, Mistral, the work that you guys did, Databricks, DBRX.
LLAMA 2、現在的 LLAMA 3、Mistral、你們所做的工作、Databricks、DBRX。
- Uh, LLaMA 2, now LLaMA 3, uh, Mistral, uh, the work that you guys did, Databricks, um, uh, DBRX.
它讓每家公司都有可能成為一家人工智能公司。
超微半導體 AMD 2024 年電腦展演講，12 分鐘濃縮重點！ (AMD's 2024 Computex Keynote: Everything Revealed in 12 Minutes)
- Now what this demo compares is the performance of Turin when running a typical enterprise deployment of LLAMA 2 virtual assistants with a minimum guaranteed latency to ensure a high quality user experience.
這個演示比較的是都靈在運行典型企業部署的 LLAMA 2 虛擬助手時的性能，並保證最低延遲，以確保高質量的用戶體驗。
- Both servers begin by loading multiple LLAMA 2 instances with each assistant being asked to summarize an uploaded document.
兩臺服務器首先都加載了多個 LLAMA 2 實例，要求每個助手對上傳的文檔進行總結。
拉丁美洲殺價祕訣大公開！🇵🇪 庫斯科旅遊購物大挑戰 (How to Bargain in Latin America | Cusco Travel Shopping Challenge)
- ¿Cómo se llama esto?
15 個鞋底。
- All right. I found an awesome sweater made from alpaca wool with a llama design on it, and it's 35 soles, which is just about $10.
好吧。 我發現了一件很棒的毛衣，是用羊駝毛做的，上面有美洲駝圖案，35索爾，也就是10美元左右。
與輝達執行長黃仁勳的對談：微軟 Build 2025 會議上的薩提亞·納德拉！ (Conversation with NVIDIA CEO Jensen Huang: Satya Nadella at Microsoft Build 2025)
- Over the course of those two years, through new algorithms, let's just say transformer, let's just say llama 70B.
在這兩年中，透過新的演算法，像是 transformer，像是 llama 70B。
- Over the course of those two years, through new algorithms, let's just say Transformer, let's just say LLaMA 70B, over the course of the last two years, we've improved the performance 40x.
而且我們願意，而且
M4 MacBook Air 統一代購機評比！哪款開發機是最終贏家？🤩 (M4 MacBook Air vs all generations | Which Dev Machine Wins?)
- I should have just went with Llama.
我們得到了 M1 的結果。
- I should have just went with Llama.
我本該選 "喇嘛 "的。
10分鐘搞懂AI競賽！ (AI Competition explained in 10 minutes)
- For example, Meta's Llama 3.1 model, which is an LLM that's 405 billion parameters in size, was released in July 2024.
舉例來說，Meta 的 Llama 3.1 模型，這是一個擁有 4050 億參數的大型語言模型，於 2024 年 7 月發布。
- So in order to do 38 septillion operations to train the Llama 3.1 model, you're going to need a lot of computing.
所以，為了進行 38 千兆兆次的運算來訓練 Llama 3.1 模型，你需要大量的運算能力。
一週生產一千萬個披薩盒！🍕 披薩的奇幻故事 (We Make 10 Million Pizza Boxes A Week | Pizza Stories)
- La pizzería se llama A Mi Manera, porque usted pide la pizza a la manera suya, y nosotros se la hacemos.
我拿著麵粉和水，只混合麵粉和水。之後，我就有了大大小小的碗
Spotify 專屬功能背後的 AI 科技大揭密！| 2025 年 12 月 18 日 (The AI-tech behind your favorite Spotify feature | December 18, 2025)
- So like for for using ChatGPT or Llama or something like that, you know, these companies scraped the Internet, grabbed as much text as they could from from Reddit conversations and forums and blog posts and news articles.
所以像是使用 ChatGPT 或 Llama 這類工具時，你知道的，這些公司會爬取網路，從 Reddit 對話、論壇、部落格文章和新聞報導中抓取盡可能多的文字。
- So for using ChatGPT or Llama or something like that, these companies scraped the Internet, grabbed as much text as they could from Reddit conversations and forums and blog posts and news articles.
它們找到的模式就像是哪些詞會跟其他哪些詞一起出現。