speculative decoding actually worth it with LM Studio

[cacheline]

I'd approach it in a boring order. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

My first guess would be this: On M3 Max, 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.

[audiodesk]

I don't think that's the whole story. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[cacheline]

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.

[latentcraft]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Mistral and compare from there. Not glamorous advic

I tested this tonight. On Threadripper box with PyTorch, I saw basically the same behavior. I expected the model choice to matter more than it did.

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