News for hackers. Stuff that matters.
Re:Tencent Releases and Open-Sources Tencent Hy4 Preview by chvid on Sunday August 30, 2026 @02:18AM (#49496141)
This is a remarkable coherent and clear reasoning trace.

Maybe you should start also comparing reasoning traces when you do your pelican benchmark.

Re:Tencent Releases and Open-Sources Tencent Hy4 Preview by armcat on Sunday August 30, 2026 @02:34AM (#49496215)
That would be very interesting but only the open models allow you to see the reasoning trace
Re:Tencent Releases and Open-Sources Tencent Hy4 Preview by kmike84 on Sunday August 30, 2026 @07:27AM (#49497744)
So, open models will be better on this benchmark, which is deserved
Re:Tencent Releases and Open-Sources Tencent Hy4 Preview by glub on Monday August 31, 2026 @01:36PM (#49512461)
For single turn prompts like this, you just have to give the model encrypted bytes and ask politely what's in them, really.

GPT-5.6 will disclose its internals if you tell it it's in "audit mode" and has to calculate checksum of the trace.