CUDA update broke my throughput

[edgecoast]

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

I'm comparing two approaches here: keep it simple on 7900 XTX, or squeeze more out of the stack and deal with the complexity.

Interested in tradeoffs more than fan loyalty.

[audiopath66]

Maybe, but I had the opposite result. I get why that seems right, but on Threadripper box the bottleneck felt elsewhere. Worth testing the simpler explanation before adopting the more interesting one.

[latentgarden]

Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[kernelmike]

Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Not

What worked for me was pretty simple. On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[dockember]

A couple things I'd check first. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[signaltrack44]

Appreciate the detail. 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.

[latentgarden]

same here on my side.

[amberpatch25]

I'd start with the simpler stack.

[kernelmike]

Appreciate the detail. I ended up in about the same place using MLX. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[audiopath66]

What worked for me was pretty simple. On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.