AWQ vs GPTQ vs GGUF right now on mini PC

[fastfield4]

Did a small comparison this afternoon. On GB200 with TensorRT-LLM, I saw basically the same behavior. I expected the model choice to matter more than it did.

If you post your exact settings, people here can probably narrow it down pretty fast.

[oxidestream]

subscribed, curious where this lands.

[sparrowkite53]

Good point. I ended up in about the same place using TensorRT-LLM.

[cudaworks]

A couple questions before I guess too hard. Are you testing this on M3 Max or something else?

My first guess would be this: On GB200, 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.

[marbletrack]

I hit almost the same issue last night.