when a plain Python stack is enough with PyTorch

[latentpeak22]

This keeps coming up in conversation so I figured I'd start a thread.

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with structured output.

Maybe obvious to some of you, but it clicked for me this week.

[plainscope50]

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.

If you post your exact settings, people here can probably narrow it down pretty fast.

[deepseeked]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals.

[plainember52]

What exact setup are you running? Is this local-only or are you routing anything externally? Turns out 'just one more test' was a lie.

[cudahouse]

Good point. The pattern around tool calling has been pretty similar for me. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[cobalttinker]

I'd approach it in a boring order. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.

[rocmriley]

I'd approach it in a boring order. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you k

A couple questions before I guess too hard. Are you testing this on 5090 or something else?

[cachebench]

Need a little more detail here. Do you see the same thing with a smaller batch or shorter prompt?

[phaseshed]

That matches what I was seeing. I ended up in about the same place using TensorRT-LLM.

[mangopeak72]

Thanks, that helps. I don't think your conclusion is far off.