dataset format recommendations

[forgemixer]

Thanks, that helps. 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.

[phasenode]

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.

[routecraft]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

[forgemixer]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

I tested this tonight. On Threadripper box with MLX, I saw basically the same behavior. I expected the model choice to matter more than it did.

[routecraft]

[quote]Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not q

Interesting, I hadn't tried that. I ended up in about the same place using SGLang. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[plainworks]

[quote][quote]Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency

I don't think that's the whole story. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model.