+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Drivers / Frameworks / Toolkits (https://www.datasculpting.ai/inference/forumdisplay.php?fid=8)
+--- Forum: Apple / MLX (https://www.datasculpting.ai/inference/forumdisplay.php?fid=33)
+--- Thread: Apple unified memory limits: setup advice (/showthread.php?tid=1022)
Pages:
1
2
Apple unified memory limits: setup advice - visionyard23 - 10-10-2026
Did a bit of testing and thought I'd post the result here.
I tried this with DeepSeek using TensorRT-LLM on 4090. The headline result was better than I expected, but NVFP4 still ended up being the limiting factor.
Happy to post more details if anyone wants them.
This is mostly for local work rather than public serving.
If I had to summarize the lesson so far, it would be that clean setup and observability beat clever guesses.
Re: Apple unified memory limits: setup advice - mellowstream - 10-12-2026
A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at quant drift. Not glamorous advice, but repeatable baselines save a lot of time.
Re: Apple unified memory limits: setup advice - plainember52 - 10-13-2026
That matches what I was seeing. I ended up in about the same place using SGLang. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: Apple unified memory limits: setup advice - visionyard23 - 10-13-2026
Good point. That matches what I saw on 4090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.
Re: Apple unified memory limits: setup advice - plainember52 - 10-13-2026
worth checking the driver first.
Re: Apple unified memory limits: setup advice - edgeriver - 10-14-2026
Ran a quick check on my box. On 4090 with SGLang, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.
Re: Apple unified memory limits: setup advice - deltagrid - 10-14-2026
A couple things I'd check first. If you're on PyTorch, 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.
Re: Apple unified memory limits: setup advice - turboyard94 - 10-14-2026
Interesting, I hadn't tried that. The pattern around router policy has been pretty similar for me. The biggest change for me was simplifying the stack instead of adding one more optimization layer.
Re: Apple unified memory limits: setup advice - torchshed - 10-14-2026
That matches what I was seeing. 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.
Re: Apple unified memory limits: setup advice - quietarch - 10-14-2026
Tried to reproduce it just now. The issue went away once I changed one setting and restarted the stack cleanly.