+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Optimization & Performance (https://www.datasculpting.ai/inference/forumdisplay.php?fid=10)
+--- Forum: Quantization (https://www.datasculpting.ai/inference/forumdisplay.php?fid=40)
+--- Thread: AWQ vs GPTQ vs GGUF right now (/showthread.php?tid=36)
AWQ vs GPTQ vs GGUF right now - cachelab - 08-27-2026
This keeps coming up in conversation so I figured I'd start a thread.
One thing I keep noticing is that people underestimate batch size until they actually try to run this stuff for more than one afternoon.
Maybe obvious to some of you, but it clicked for me this week.
Re: AWQ vs GPTQ vs GGUF right now - cudaline - 08-27-2026
What worked for me was pretty simple. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives.
Re: AWQ vs GPTQ vs GGUF right now - torchshed - 08-27-2026
Not sure I agree there. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Worth testing the simpler explanation before adopting the more interesting one.
Re: AWQ vs GPTQ vs GGUF right now - paperrunner - 08-27-2026
A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: AWQ vs GPTQ vs GGUF right now - vramscope - 08-27-2026
That matches what I was seeing. The pattern around small local stacks has been pretty similar for me.
Re: AWQ vs GPTQ vs GGUF right now - southtinker - 08-27-2026
My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at small local stacks. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.
Re: AWQ vs GPTQ vs GGUF right now - cudaline - 08-28-2026
My first guess would be this: If you're on PyTorch, I'd confirm the version first and then reduce variables until the behavior is obvious.
Re: AWQ vs GPTQ vs GGUF right now - marbletrack - 08-28-2026
Quote:My first guess would be this: If you're on PyTorch, I'd confirm the version first and then reduce variables until the behavior is obvious.
that fixed it for me too.
Re: AWQ vs GPTQ vs GGUF right now - fusedtinker - 08-28-2026
Quote:[quote]My first guess would be this: If you're on PyTorch, I'd confirm the version first and then reduce variables until the behavior is obvious.[/quo
Can you fill in a bit more context? Is this local-only or are you routing anything externally?
Re: AWQ vs GPTQ vs GGUF right now - turboscope - 08-29-2026
Quote:[quote][quote]My first guess would be this: If you're on PyTorch, I'd confirm the version first and then reduce variables until the behavior is obviou
Ran a quick check on my box. The issue went away once I changed one setting and restarted the stack cleanly. The surprise was that memory pressure showed up before throughput collapsed. Turns out 'just one more test' was a lie.