GPT-5.4

Post by daft_pink » Thu, Mar 05, 2026, 8:23 PM UTC

Post by ozgung » Thu, Mar 05, 2026, 8:23 PM UTC

Chance-Device wrote: post #13

I’m sure the military and security services will enjoy it.

Post by MattDaEskimo » Thu, Mar 05, 2026, 8:26 PM UTC

mattas wrote: post #5

"GPT‑5.4 interprets screenshots of a browser interface and interacts with UI elements through coordinate-based clicking to send emails and schedule a calendar event." They show an example of 5.4 clicking around in Gmail to send an email. I still think this is the wrong interface to be interacting with the internet. Why not use Gmail APIs? No need to do any screenshot interpretation or coordinate-based clicking.

Post by patcon » Thu, Mar 05, 2026, 8:26 PM UTC

skilltissue wrote: post #273

elmean wrote: post #195

[flagged]

Don't use the site this way. https://news.ycombinator.com/newsguidelines.html

Post by woadwarrior01 » Thu, Mar 05, 2026, 8:26 PM UTC

sillysaurusx wrote: post #251

tedsanders wrote: post #59

Earlier quoted context omitted.

Yeah, long context vs compaction is always an interesting tradeoff. More information isn't always better for LLMs, as each token adds distraction, cost, and latency. There's no single optimum for all use cases. For Codex, we're making 1M context experimentally available, but we're not making it the default experience for everyone, as from our testing we think that shorter context plus compaction works best for most p…

You may want to look over this thread from cperciva: https://x.com/cperciva/status/2029645027358495156 I too tried Codex and found it similarly hard to control over long contexts. It ended up coding an app that spit out millions of tiny files which were technically smaller than the original files it was supposed to optimize, except due to there being millions of them, actual hard drive usage was 18x larger. It seemed…