video model setup pain points for Phi

[bytelab]

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

It feels like the conversation has shifted hard toward inference and serving details lately, and honestly that makes sense once you start dealing with router policy.

Curious whether that lines up with what other people are seeing.

This is mostly for local work rather than public serving.

[torchshed]

that fixed it for me too.

[signalcraft14]

Good point. The pattern around HBM constraints 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.

[embershed99]

What worked for me was pretty simple. On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[tensorworks]

What worked for me was pretty simple. On mini PC, the easiest win is often to back off the aggressive settings and see whether the problem survives. O

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with Mistral and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[mellowstream]

I think the bottleneck is somewhere else. I get why that seems right, but on A6000 the bottleneck felt elsewhere. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[phasetrack]

What exact setup are you running? Is this local-only or are you routing anything externally?

[kernelhouse70]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tokens/sec. Once you know the plain path works, then add the faster path back in piece by piece.

[thetaplane]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tokens/sec. Once you kno

I tested this tonight. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

[fastbench]

I'd approach it in a boring order. If you're on MLX, 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.