Hacker News: ameddserMhttps://news.ycombinator.com/user?id=ameddserMHacker News RSShttps://hnrss.org/hnrss v2.1.1Thu, 03 Sep 2026 07:27:27 +0000<![CDATA[New comment by ameddserM in "Fable 5.1 World Modeling"]]>This looks great.

]]>Wed, 02 Sep 2026 23:00:05 +0000https://news.ycombinator.com/item?id=49543762ameddserMhttps://news.ycombinator.com/item?id=49543762https://news.ycombinator.com/item?id=49543762<![CDATA[New comment by ameddserM in "The first benchmark to test AI agent's video editing capability"]]>AI agents can write code. AI agents can browse the web. AI agents can pass the bar exam. Can they edit a video? We gave the 7 best frontier models 100 real-world post-production tasks, scored by 20 industry experts. Best agent: barely crosses 30% Human experts: 89% Introducing AgenticVBench.

]]>Sat, 23 May 2026 17:12:42 +0000https://news.ycombinator.com/item?id=48249338ameddserMhttps://news.ycombinator.com/item?id=48249338https://news.ycombinator.com/item?id=48249338<![CDATA[The first benchmark to test AI agent's video editing capability]]>Article URL: https://agenticvbench.com/

Comments URL: https://news.ycombinator.com/item?id=48249337

Points: 2

# Comments: 1

]]>Sat, 23 May 2026 17:12:42 +0000https://agenticvbench.com/ameddserMhttps://news.ycombinator.com/item?id=48249337https://news.ycombinator.com/item?id=48249337