Used A6000 worth it now

[rocmriley]

This keeps coming up in conversation so I figured I'd start a thread.

One thing I keep noticing is that people underestimate prefix caching until they actually try to run this stuff for more than one afternoon.

Wondering if this is just where the whole space is heading now.

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

[vrambandit]

My first guess would be this: On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

[dockkite69]

I'd approach it in a boring order. On A6000, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

[torchshed]

Good point. That matches what I saw on mini PC. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[signaltrack44]

What worked for me was pretty simple. On GB200, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[rocmriley]

Maybe, but I had the opposite result. I've seen people blame consumer GPUs when the real issue was much more mundane. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[thetarunner]

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. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[cudapoint]

Tried to reproduce it just now. I couldn't reproduce the slowdown until I pushed context much higher. The surprise was that memory pressure showed up before throughput collapsed.

[vrambandit]

Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?

[thetarunner]

I'd approach it in a boring order. On 7900 XTX, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

I'm happy to test one more variant if you want a second data point.