HIP kernel compile issue: worth it?

[macropoint]

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

The more I work with local setups, the more it seems like small local stacks decides everything downstream.

Curious whether that lines up with what other people are seeing.

[rocmriley]

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.

[quantshed]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Kimi and compare from there.

[northhall88]

What worked for me was pretty simple. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious.

[tensorworks]

Interesting, I hadn't tried that. I ended up in about the same place using TensorRT-LLM. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[rocmriley]

Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?

[turboyard94]

I'd approach it in a boring order. If you're on LM Studio, 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.

[mintquill]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. I expected the model choice to matter more than it did.

[tensorworks]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. I expected the model choice to matter more than it d

Maybe, but I had the opposite result. I've seen people blame VRAM headroom when the real issue was much more mundane. Worth testing the simpler explanation before adopting the more interesting one.

[cozykite]

following this because I'm in the same boat.