dedicated AI gadgets you actually use: worth it?

[flintriver50]

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 speculative decoding.

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

[bytepilot]

That matches what I was seeing. I don't think your conclusion is far off. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[audiorunner74]

My first guess would be this: 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.

If you post your exact settings, people here can probably narrow it down pretty fast.

[latentscope35]

What exact setup are you running? Do you see the same thing with a smaller batch or shorter prompt?

[routeline23]

Ran a quick check on my box. My result was mixed: first run looked bad, second run looked normal. The surprise was that memory pressure showed up before throughput collapsed.

[routecrate]

Appreciate the detail. I don't think your conclusion is far off.

[rasterbench]

that fixed it for me too.

[bytepilot]

I tested this tonight. My result was mixed: first run looked bad, second run looked normal. The main thing that moved was latency, not quality.

[routecrate]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with Phi and compare from there.