According to the DrivingBench report, how did OpenAI’s GPT-6View Astra become the first commercial large language model to steer a real carAI-generated editorial illustration; not footage from the DrivingBench test.
AI 提示
Create a landscape editorial hero image for this Studio Global article: According to the DrivingBench report, how did OpenAI’s GPT-6View Astra become the first commercial large language model to steer a real car. Article summary: DrivingBench’s result was a narrow but notable proof of concept: OpenAI’s GPT-6 Astra—not “GPT-6View Astra” in the available sources—was the only tested model to complete a real-car cone course, succeeding on its second . Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fak
openai.com
一個通用 AI 模型真係控制咗一架車,仲行完指定路線。不過,行得完測試路線,同可以放手畀佢上路,係兩回事。在 DrivingBench 的四模型比較中,OpenAI 的 GPT-6 Astra(並非「GPT-6View Astra」)是唯一完成停車場雪糕筒路線的模型;它首次嘗試未能完成,要到第二次才成功。1415
它點樣控制架車?
Tobias Gessler、Simon Mahns 和 Aditya Ramabadran 設計 DrivingBench,是要測試通用 AI 模型能否處理真實世界的駕駛任務,而不只是在模擬環境中表現良好。測試使用一架 Toyota Corolla,經 Comma 4 硬件及建基於 openpilot 的控制程式連接模型。每個模型最多可在同一段持續對話中嘗試三次,利用先前的結果調整做法。1136