News for hackers. Stuff that matters.
Re:Every AI Subscription Is a Ticking Time Bomb for Enterprise by dools on Sunday May 17, 2026 @12:28PM (#48170398)
Kimi k2.6 is about on par with GPT 5.2 so I’d say open weight models are about 6 months behind.
Re:Every AI Subscription Is a Ticking Time Bomb for Enterprise by cbg0 on Sunday May 17, 2026 @12:57PM (#48170679)
The Q4 quantization requires about 600GB of RAM without context, not exactly consumer hardware friendly.
Re:Every AI Subscription Is a Ticking Time Bomb for Enterprise by janderland on Sunday May 17, 2026 @12:54PM (#48170653)
Has Kimi found a way to vastly reduce the amount of VRAM required without running at 3 tokens per second? That’s the real concern.
Re:Every AI Subscription Is a Ticking Time Bomb for Enterprise by dools on Sunday May 17, 2026 @07:56PM (#48174225)
I said "open weight" rather than "local". I mean, local if you have $240k to drop on GPUs but you can run Kimi k2.6 on a B300 cluster for ~$50/hour too.