sandboxing agent actions

[batchwire]

Curious what people here are seeing.

I'm trying to decide whether to keep the current stack or change course. Main question is how people here are handling NVFP4.

If you've tested something similar, what did you keep and what did you stop bothering with? At this point the real bottleneck is my patience.

[routeline23]

Curious what people here are seeing.

A couple things I'd check first. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.

[dockridge]

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

[batchscope72]

Tried to reproduce it just now. My result was mixed: first run looked bad, second run looked normal.

[thetaplane]

worth checking the driver first.

[cozywire]

My first guess would be this: On Threadripper box, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[stackpilot]

Good point. 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.

[turboscope]

worth checking the driver first.

[southden]

worth checking the driver first.

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with FLUX and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[sanevault]

worth checking the driver first.

Did a small comparison this afternoon. On 5090 with Ollama, I saw basically the same behavior. I expected the model choice to matter more than it did.