DeepSeek V4.1 impressions

[routecraft]

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 tokens/sec decides everything downstream.

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

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs.

[routecraft]

My first guess would be this: On 5090, 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.

[edgecoast]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

[batchpeak]

Ran a quick check on my box. I couldn't reproduce the slowdown until I pushed context much higher. The main thing that moved was latency, not quality.

Did a small comparison this afternoon. The issue went away once I changed one setting and restarted the stack cleanly. The surprise was that memory pressure showed up before throughput collapsed.

[routecraft]

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. Not glamorous advice, but repeatable baselines save a lot of time.

[tensorport]

Good point. The pattern around tool calling has been pretty similar for me.

[routecraft]

Good point. The pattern around tool calling has been pretty similar for me.

What worked for me was pretty simple. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at MoE serving. Not glamorous advice, but repeatable baselines save a lot of time.

[latentpeak22]

A couple things I'd check first. On B200, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[routecraft]

Ran a quick check on my box. The issue went away once I changed one setting and restarted the stack cleanly. I expected the model choice to matter more than it did.

[rocmhouse90]

Did a small comparison this afternoon. I couldn't reproduce the slowdown until I pushed context much higher. I expected the model choice to matter more than it did.