system prompt habits that helped on M3 Max

[cozywire]

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 continuous batching decides everything downstream.

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

[paperport64]

worth checking the driver first.

[thetafield]

My first guess would be this: On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives. Not glamorous advice, but repeatable baselines save a lot of time.

[mlxgarden]

A couple things I'd check first. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[batchscope72]

A couple questions before I guess too hard. Do you see the same thing with a smaller batch or shorter prompt?

[fusedhall56]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at embeddings. Not glamorous advice, but repeatable baselines save a lot of time.

[thetafield]

Appreciate the detail. That matches what I saw on 4090. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[northgarden37]

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.

[rasterhall]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[fusedhall56]

My first guess would be this: If you're on vLLM, 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.