New theme updated, we hope you like it

[Data Sculpting]

We pushed a new theme update today.

The goal is a cleaner dark layout that fits the Data Sculpting branding and feels easier to read across desktop and mobile. If you notice anything broken or misaligned, please mention it in Community Announcements.

[cudacrate]

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. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[Data Sculpting]

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

[Data Sculpting]

What exact setup are you running? Are you testing this on 7900 XTX or something else?

[sparrowhall]

A couple things I'd check first. If you're on MLX, I'd confirm the version first and then reduce variables until the behavior is obvious.

[Data Sculpting]

What worked for me was pretty simple. 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.

[Data Sculpting]

I think the bottleneck is somewhere else. I get why that seems right, but on 5090 the bottleneck felt elsewhere. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[batchsize]

I think the bottleneck is somewhere else. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Worth testing the simpler explanation before adopting the more interesting one.

[Data Sculpting]

that fixed it for me too.

[dockbrook21]

Need a little more detail here. Is this local-only or are you routing anything externally?