NVIDIA driver regression: early impressions

[kernelmike]

Did a bit of testing and thought I'd post the result here.

Short version: PyTorch behaved well for me, but only after I stopped ignoring token cost.

Happy to post more details if anyone wants them.

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.

[thetapath]

Appreciate the detail. I ended up in about the same place using PyTorch.

[audioriver]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[torchshed66]

Can you fill in a bit more context? Are you testing this on B200 or something else?

[northcraft90]

Interesting, I hadn't tried that. That matches what I saw on 7900 XTX. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[cobaltspark]

Did a small comparison this afternoon. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.

[phaseport36]

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. Not glamorous advice, but repeatable baselines save a lot of time.

[mintforge]

Thanks, that helps. That matches what I saw on dual 3090.

[amberpatch25]

Thanks, that helps. That matches what I saw on dual 3090.

Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly. The main thing that moved was latency, not quality.

[mangopeak72]

subscribed, curious where this lands.