prefill/decode separation in practice

[mangofield33]

Appreciate the detail. That matches what I saw on M3 Max. I got better results once I stopped chasing one metric and just looked at whether the workfl

I'd start with the simpler stack.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.

[cudapoint]

My first guess would be this: On 7900 XTX, 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.

[oxidemixer]

My first guess would be this: On 7900 XTX, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glam

What worked for me was pretty simple. 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.

[audioforge]

I'd start with the simpler stack.

[cacheline]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with FLUX and compare from there. Once you know the plain path works, then add the faster path back in piece by piece.