reranking still worth it for small sets: early impressions

[phasespark]

Did a bit of testing and thought I'd post the result here.

I tried this with Kimi using vLLM on 7900 XTX. The headline result was better than I expected, but structured output still ended up being the limiting factor.

Happy to post more details if anyone wants them.

[cozysmith]

Appreciate the detail. I don't think your conclusion is far off.

[thetaember38]

A couple things I'd check first. On dual 3090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[batchsize]

That matches what I was seeing. The pattern around continuous batching 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.

[sanepeak48]

A couple questions before I guess too hard. Is this local-only or are you routing anything externally?

[deepseeked]

A couple questions before I guess too hard. Is this local-only or are you routing anything externally?

A couple questions before I guess too hard. Do you see the same thing with a smaller batch or shorter prompt?

[latentcraft]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Llama and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[turbogrid11]

Appreciate the detail. I don't think your conclusion is far off.

[prompttinker44]

Appreciate the detail. That matches what I saw on GB200. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[fusedriver]

What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?