Please keep off topic stuff in the Off Topic forum

[Data Sculpting]

Quick reminder: please keep off-topic conversations in the Off Topic forum.

That helps the technical sections stay useful for people searching for hardware, models, engines, and troubleshooting threads later.

[vramstream]

Need a little more detail here. Is this local-only or are you routing anything externally?

[turboscope]

Can you fill in a bit more context? Are you testing this on 5090 or something else?

[Data Sculpting]

following this because I'm in the same boat.

[vramghost]

Appreciate the detail. The pattern around prefill/decode split 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.

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

[Data Sculpting]

What worked for me was pretty simple. On 7900 XTX, 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.

[deltapoint17]

Interesting, I hadn't tried that. I don't think your conclusion is far off.

[batchyard]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[Data Sculpting]

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

[Data Sculpting]

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.