driver/framework news roundup on dual 3090

[fastlab]

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

This isn't revolutionary, but the combination of SGLang and A6000 felt surprisingly usable once I tuned around MoE serving.

If someone wants the exact settings I can clean them up and post them.

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

[circuitkite]

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

Appreciate the detail. That matches what I saw on 5090.

[thetashed24]

I hit almost the same issue last night.

[bytestack]

worth checking the driver first.

[mangowire]

A couple things I'd check first. If you're on Ollama, I'd confirm the version first and then reduce variables until the behavior is obvious.

[audiopoint18]

That matches what I was seeing. I ended up in about the same place using PyTorch. The biggest change for me was simplifying the stack instead of adding one more optimization layer.