Inference › Drivers / Frameworks / Toolkits › AMD / ROCm v
Instinct pricing thoughts

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
Instinct pricing thoughts
papermixer4 Offline
Junior Member
**
Posts: 44
Threads: 4
Joined: Aug 2026
Reputation: 0
#3
10-03-2026, 11:20 AM
Quote:I tested this tonight. On GB200 with PyTorch, I saw basically the same behavior. The main thing that moved was latency, not quality.

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Phi and compare from there.
Find
Reply


Messages In This Thread
Instinct pricing thoughts - by cacheline - 10-01-2026, 05:59 PM
Re: Instinct pricing thoughts - by rocmriley - 10-03-2026, 06:40 AM
Re: Instinct pricing thoughts - by cozywire - 10-03-2026, 01:05 PM
Re: Instinct pricing thoughts - by papermixer4 - 10-03-2026, 11:20 AM
Re: Instinct pricing thoughts - by edgeriver - 10-04-2026, 06:14 AM
Re: Instinct pricing thoughts - by relaynest82 - 10-03-2026, 03:52 PM
Re: Instinct pricing thoughts - by rocmriley - 10-07-2026, 09:10 AM
Re: Instinct pricing thoughts - by fastquill74 - 10-03-2026, 04:46 PM
Re: Instinct pricing thoughts - by mangofield33 - 10-03-2026, 05:30 PM
Re: Instinct pricing thoughts - by batchyard - 10-04-2026, 06:56 AM
Re: Instinct pricing thoughts - by quantshed - 10-07-2026, 07:22 AM
Re: Instinct pricing thoughts - by phasetrack - 10-07-2026, 08:39 AM
Re: Instinct pricing thoughts - by macropoint - 10-07-2026, 10:01 PM
Re: Instinct pricing thoughts - by northhall88 - 10-09-2026, 10:42 AM
Re: Instinct pricing thoughts - by papermixer4 - 10-09-2026, 03:04 PM

Forum Jump:


Users browsing this thread: 1 Guest(s)