X でみかけた LLM のコンテキストを圧縮するっていう OSS、現時点では環境変数設定すれば off にはできそうだけどテレメトリを外部に送信する機能があるし https://t.co/rOhsnX625K 仕組み的にフックで容易に機密情報抜く実装に変更できるだろうし、さすがに怖い。こういうの増えそうだなあ・・
We heard you wanted more examples of Cinematic Video Overviews. Here's a few of our faves, created by the most talented, creative, intelligent people we know (you!)
Please keep sharing your example videos and prompts. We love to see them 🥰 🧵
Veo and Gemini image models such as Nano Banana are now live in the OpenAI compatibility layer!
Test our models in your existing Python/JS pipelines with zero SDK rewrite. Just update exactly 3 lines of code (𝚊𝚙𝚒_𝚔𝚎𝚢, 𝚋𝚊𝚜𝚎_𝚞𝚛𝚕, 𝚖𝚘𝚍𝚎𝚕).
Using GKE Autopilot and Compute Classes is about to get a whole lot faster with concurrent node-pool creation!
When you use these features, GKE creates a set of node pools under the hood for different machine shapes--an action that was historically a key scaling bottleneck. No longer. https://t.co/sJk3hX8Qi7