JAX for inference only: what are you seeing?

[edgecoast]

I think the bottleneck is somewhere else. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model. Could be a workload difference, but I wouldn't generalize too fast from one setup.

[torchshed]

I think the bottleneck is somewhere else. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the model.

That matches what I was seeing. I ended up in about the same place using LM Studio. The biggest change for me was simplifying the stack instead of adding one more optimization layer.

[rasterhall]

[quote]I think the bottleneck is somewhere else. I wouldn't optimize around that first. On my side the bigger swing came from the runtime and not the

What worked for me was pretty simple. If you're on SGLang, 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.

[prompttinker44]

Need a little more detail here. What model, what runtime, and roughly what context size?

[edgequill]

Appreciate the detail. 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.

[torchdesk]

What worked for me was pretty simple. On 4090, 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.