Inference
when video is still not worth the wait for DeepSeek - Printable Version

+- Inference (https://www.datasculpting.ai/inference)
+-- Forum: Evaluation & Testing (https://www.datasculpting.ai/inference/forumdisplay.php?fid=16)
+--- Forum: Quality / Reliability Testing (https://www.datasculpting.ai/inference/forumdisplay.php?fid=68)
+--- Thread: when video is still not worth the wait for DeepSeek (/showthread.php?tid=302)

Pages: 1 2


when video is still not worth the wait for DeepSeek - turboscope - 09-15-2026

Posting this in case it saves someone else a little time.

I'm helping a small team set up a local workflow. We don't need a giant stack, but we do care about MoE serving.

Would rather learn from your scars than collect my own all at once.


Re: when video is still not worth the wait for DeepSeek - quantnest - 09-15-2026

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


Re: when video is still not worth the wait for DeepSeek - bytestack - 09-17-2026

Maybe, but I had the opposite result. I've seen people blame speculative decoding when the real issue was much more mundane.


Re: when video is still not worth the wait for DeepSeek - quiettoken - 09-17-2026

I'd start with the simpler stack.


Re: when video is still not worth the wait for DeepSeek - mangovault67 - 09-17-2026

I tested this tonight. I couldn't reproduce the slowdown until I pushed context much higher. The surprise was that memory pressure showed up before throughput collapsed.


Re: when video is still not worth the wait for DeepSeek - fastfield4 - 09-18-2026

Need a little more detail here. Do you see the same thing with a smaller batch or shorter prompt?

If you post your exact settings, people here can probably narrow it down pretty fast.


Re: when video is still not worth the wait for DeepSeek - batchscope72 - 09-19-2026

What worked for me was pretty simple. On M3 Max, the easiest win is often to back off the aggressive settings and see whether the problem survives.


Re: when video is still not worth the wait for DeepSeek - byteyard - 09-22-2026

Appreciate the detail. I ended up in about the same place using vLLM.


Re: when video is still not worth the wait for DeepSeek - southtinker - 09-24-2026

Quote:Appreciate the detail. I ended up in about the same place using vLLM.

I hit almost the same issue last night.


Re: when video is still not worth the wait for DeepSeek - macrotinker - 09-24-2026

I'd start with the simpler stack.