do you even need a vector DB here: setup advice

[rocmriley]

Maybe, but I had the opposite result. I get why that seems right, but on 4090 the bottleneck felt elsewhere. Worth testing the simpler explanation before adopting the more interesting one.

[quantyard]

Thanks, that helps. I don't think your conclusion is far off. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[cacheline]

Thanks, that helps. That matches what I saw on 5090. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[torchsupply99]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[sparrowghost]

same here on my side.

[flintriver50]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at workstation thermals.