Help with Embeddings: what are you seeing?

[thetanest]

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

This isn't revolutionary, but the combination of TensorRT-LLM and Threadripper box felt surprisingly usable once I tuned around consumer GPUs.

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

[ggufwire22]

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

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tool calling. Once you know the plain path works, then add the faster path back in piece by piece.

[fusedhall]

Appreciate the detail. I ended up in about the same place using PyTorch. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[edgecoast]

Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[labline]

Maybe, but I had the opposite result. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Coul

Good point. The pattern around INT4 has been pretty similar for me. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[vramscope]

I'd approach it in a boring order. On GB200, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[dockriver]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at VRAM headroom. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[vectorshed43]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at VRAM headroom. A lot of

subscribed, curious where this lands.

[ggufwire22]

Appreciate the detail. I ended up in about the same place using PyTorch. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.