Search / Retrieval / Reranking discussion

[rasterbench]

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

I've gone back and forth between PyTorch and TensorRT-LLM for Mixtral. The interesting difference for me has been structured output.

Interested in tradeoffs more than fan loyalty.

[localtinker]

Appreciate the detail. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[batchwire]

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.

[phasespark]

A couple things I'd check first. On 7900 XTX, 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.

[batchsize]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. I expected the model choice to matter more than it did.

[vrampilot]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.

[routeline23]

I'd approach it in a boring order. If you're on llama.cpp, 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.

[dockriver]

I'd approach it in a boring order. If you're on llama.cpp, I'd confirm the version first and then reduce variables until the behavior is obvious. Once

that fixed it for me too.

[cacheline]

I tested this tonight. On 5090 with LM Studio, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.

[benchplease]

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