Inference › Drivers / Frameworks / Toolkits › Apple / MLX v
MLX model loading quirks

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Linear Mode
MLX model loading quirks
rocmrunner40 Offline
Member
***
Posts: 57
Threads: 8
Joined: Oct 2026
Reputation: 0
#1
10-17-2026, 02:20 AM
Has anyone else run into this?

I'm mostly working with DeepSeek in llama.cpp on Threadripper box. The part I'm trying to understand is batch size.

Not looking for the perfect answer, mostly what worked in real use.
Find
Reply


Messages In This Thread
MLX model loading quirks - by rocmrunner40 - 10-17-2026, 02:20 AM
Re: MLX model loading quirks - by pixelcrate82 - 10-17-2026, 02:49 PM
Re: MLX model loading quirks - by labnest - 10-19-2026, 12:28 AM
Re: MLX model loading quirks - by sanescope - 10-19-2026, 05:37 AM
Re: MLX model loading quirks - by plainember52 - 10-19-2026, 12:57 AM
Re: MLX model loading quirks - by torchshed - 10-19-2026, 01:40 AM
Re: MLX model loading quirks - by thetaarch43 - 10-18-2026, 10:47 PM
Re: MLX model loading quirks - by dockridge - 10-19-2026, 02:36 AM
Re: MLX model loading quirks - by mlxgarden - 10-19-2026, 03:30 AM
Re: MLX model loading quirks - by quietarch - 10-19-2026, 12:14 AM
Re: MLX model loading quirks - by sanescope - 10-19-2026, 04:19 AM
Re: MLX model loading quirks - by visionyard23 - 10-19-2026, 05:21 AM

Forum Jump:


Users browsing this thread: 2 Guest(s)