dual GPU split inference

[dockkite69]

Interesting, I hadn't tried that. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[turboyard94]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[embersmith]

Interesting, I hadn't tried that. That matches what I saw on A6000. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[flintriver50]

What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?

Not sure I agree there. I've seen people blame structured output when the real issue was much more mundane.

[rocmriley]

What worked for me was pretty simple. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[sanevault]

What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?

[batchember]

Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly.

[dockworks23]

Spent some time on this today and wanted to compare notes.

I'm comparing two approaches here: keep it simple on A6000, or squeeze more out of the stack and deal with the complexity.

What would you pick if you had to live with it for a few months?

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.

[torchshed]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at INT4. Not glamorous advice, but repeatable baselines save a lot of time.

[rocmriley]

Interesting, I hadn't tried that. That matches what I saw on A6000. I got better results once I stopped chasing one metric and just looked at whether

I'd approach it in a boring order. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.