70B still worth the pain

[fastbench]

A couple things I'd check first. Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[amberkite]

I'd approach it in a boring order. On 4090, 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.

[routecrate]

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

[fastscope]

worth checking the driver first.

[turboscope]

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 NVFP4 decides everything downstream.

Maybe obvious to some of you, but it clicked for me this week.

[ggufwire22]

Thanks, that helps. 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.

[mlxgarden]

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

[batchsize]

That matches what I was seeing. 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.

[turboscope5]

I'd approach it in a boring order. Before rebuilding anything, I'd test one clean baseline with FLUX and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[quietsupply11]

I'd start with the simpler stack.