AWQ vs GPTQ vs GGUF right now

[plainline]

Thought I'd post a small recipe because a lot of threads ask similar things.

Maybe useful as a starting point: vLLM on 4090, keeping speculative decoding in mind from the start.

Feel free to tear it apart if you see something dumb.

I care more about stability than squeezing the last possible tokens/sec out of it.

[bytehall]

My first guess would be this: On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[fastbench]

Good point. I ended up in about the same place using MLX. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[cudaline]

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals.

[phasestream53]

Appreciate the detail. That matches what I saw on dual 3090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[cudaline]

Need a little more detail here. Is this local-only or are you routing anything externally?

[fastbench]

I'd approach it in a boring order. If you're on LM Studio, 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.

[audionode8]

Thanks, that helps. The pattern around small local stacks has been pretty similar for me. Turns out 'just one more test' was a lie.

[relaynest82]

subscribed, curious where this lands.

[sparrowgrid]

subscribed, curious where this lands.

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at prefill/decode split. Not glamorous advice, but repeatable baselines save a lot of time.