prompt length getting silly

[fusedhall56]

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

I'm comparing two approaches here: keep it simple on 4090, or squeeze more out of the stack and deal with the complexity.

What would you pick if you had to live with it for a few months?

This is mostly for local work rather than public serving.

[torchshed66]

My first guess would be this: 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.

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

[audioforge]

Thanks, that helps. The pattern around tail latency has been pretty similar for me. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[northgarden37]

worth checking the driver first.

[mintstation99]

A couple things I'd check first. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Not glamorous advice, but repeatable baselines save a lot of time.

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

[promptspark58]

A couple things I'd check first. If you're on MLX, 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.

[mintstation99]

following this because I'm in the same boat.

[paperport64]

My first guess would be this: If you're on Ollama, 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.

[nanopatch]

A couple questions before I guess too hard. What model, what runtime, and roughly what context size?

[localpatch]

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.