where reranking mattered

[audionode8]

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

After a few runs on M3 Max, I ended up with decent performance, but the setup only really made sense once I adjusted around long context.

Curious if anyone else landed in the same place.

[quietsupply11]

Can you fill in a bit more context? Are you testing this on GB200 or something else?

[circuitworks90]

My first guess would be this: 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.

[northgarden37]

I'd start with the simpler stack.

[modrunner27]

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.

[northhall88]

I don't think that's the whole story. I've seen people blame router policy when the real issue was much more mundane.

[modrunner27]

I'd start with the simpler stack.

[quietsupply11]

I'd start with the simpler stack.

I'm happy to test one more variant if you want a second data point.

[quietsupply11]

A couple things I'd check first. On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

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

[griffinloom66]

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