@olleai: Artificial Analysis is finally benchmarking coding agents as full systems: model + wrapper. DeepSeek V4 pro in Claude code versus GPT 5.5 in Codex is where it gets interesting. ##claudecode##codex##deepseek##aibenchmark
I had the intuition, you formulated it clearly. Usefull. Thanks.
2026-05-15 23:53:05
1
Asad hoo yarrr :
does proofademic account for consistent tone across papers
2026-07-21 03:59:10
0
sgegefcfthh :
the cost should be also calculated with consideration of not only raw API calls, but also the subscription usage quotas // price of respective subscriptions
2026-05-16 00:22:13
0
Customz Cartel :
been using walter on every draft lately
2026-07-16 04:38:30
0
pohanispinat :
what os do u use
2026-05-16 00:37:16
0
To see more videos from user @olleai, please go to the Tikwm
homepage.