clean API wrappers for local agents for Llama

[cobalttinker]

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

The more I work with local setups, the more it seems like HBM constraints decides everything downstream.

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

For context, I'm not chasing leaderboard numbers. I mostly want the system to feel consistent day to day.

[mlxgarden]

Tried to reproduce it just now. I couldn't reproduce the slowdown until I pushed context much higher.

[kernelhouse70]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Gemma and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[torchkite]

that fixed it for me too.

I'm happy to test one more variant if you want a second data point.

[mlxgarden]

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

[rocmrunner40]

I'd approach it in a boring order. On dual 3090, the easiest win is often to back off the aggressive settings and see whether the problem survives. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[southpatch]

Thanks, that helps. That matches what I saw on A6000. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable. Turns out 'just one more test' was a lie.

[rocmstack25]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with Mixtral and compare from there.

[phaseport36]

Good point. The pattern around embeddings 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.

[oxiderunner]

Good point. 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.

I'm happy to test one more variant if you want a second data point.