adapter stacking for DeepSeek

[griffinloom66]

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

[cudascope]

I'd start with the simpler stack.

[circuitworks90]

Appreciate the detail. I ended up in about the same place using MLX. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[cudascope]

What worked for me was pretty simple. 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.

[batchsize]

What worked for me was pretty simple. On A6000, 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.