prompt length getting silly

[localpatch]

Posting this in case it saves someone else a little time.

If you were starting from scratch today for this use case, what would you choose first and what would you ignore?

Would rather learn from your scars than collect my own all at once.

[nanopatch]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there.

[thetaarch43]

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

[paperport64]

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. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[thetafield]

A couple things I'd check first. On Threadripper box, 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.

[quiettoken]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at continuous batching. Not glamorous advice, but repeatable baselines save a lot of time.

[northgarden37]

worth checking the driver first.

[fusedhall56]

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

[thetaarch43]

I tested this tonight. I couldn't reproduce the slowdown until I pushed context much higher.

[thetafield]

My first guess would be this: On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of

worth checking the driver first.