DeepSeek V4.1 impressions with PyTorch

[routecraft]

Thought I'd post a small recipe because a lot of threads ask similar things.

Just sharing a setup that ended up working reasonably well: DeepSeek, SGLang, M3 Max, with a lot of attention on batch size.

Feel free to tear it apart if you see something dumb.

This is mostly for local work rather than public serving.

[routecraft]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tail latency. Once you know the plain path works, then add the faster path back in piece by piece.

[latentgarden]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tail latency. Not glamorous advice, but repeatable baselines save a lot of time.

[routecraft]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. I expected the model choice to matter more than it did.

[macrolab]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. I expected the model choice to matter more than it

That matches what I was seeing. That matches what I saw on M3 Max. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[routecraft]

Can you fill in a bit more context? Do you see the same thing with a smaller batch or shorter prompt?

[routecraft]

Need a little more detail here. Are you testing this on Threadripper box or something else?

[rocmriley]

following this because I'm in the same boat.

[routecraft]

I'd start with the simpler stack.

[torchsupply99]

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