+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Drivers / Frameworks / Toolkits (https://www.datasculpting.ai/inference/forumdisplay.php?fid=8)
+--- Forum: PyTorch / JAX / Other Frameworks (https://www.datasculpting.ai/inference/forumdisplay.php?fid=34)
+--- Thread: PyTorch compile got faster? (/showthread.php?tid=715)
Pages:
1
2
PyTorch compile got faster? - cachebench - 10-03-2026
This keeps coming up in conversation so I figured I'd start a thread.
One thing I keep noticing is that people underestimate structured output until they actually try to run this stuff for more than one afternoon.
Maybe obvious to some of you, but it clicked for me this week.
Re: PyTorch compile got faster? - northgarden37 - 10-03-2026
I'd start with the simpler stack.
Re: PyTorch compile got faster? - phasestream53 - 10-03-2026
I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at NVFP4. Not glamorous advice, but repeatable baselines save a lot of time.
Re: PyTorch compile got faster? - northgrid33 - 10-04-2026
What worked for me was pretty simple. If you're on LM Studio, I'd confirm the version first and then reduce variables until the behavior is obvious.
Re: PyTorch compile got faster? - deepseeked - 10-05-2026
Thanks, that helps. The pattern around token cost has been pretty similar for me.
Re: PyTorch compile got faster? - thetarunner - 10-05-2026
What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at MoE serving. Not glamorous advice, but repeatable baselines save a lot of time.
Re: PyTorch compile got faster? - labpath - 10-05-2026
Good point. That matches what I saw on GB200.
Re: PyTorch compile got faster? - routecraft - 10-05-2026
Quote:Good point. That matches what I saw on GB200.
My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at long context. Once you know the plain path works, then add the faster path back in piece by piece.
Re: PyTorch compile got faster? - labpath - 10-05-2026
Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly.
Re: PyTorch compile got faster? - rocmpath - 10-05-2026
I'd approach it in a boring order. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.