when RAG was overkill for DeepSeek

[quantshed]

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 prefill/decode split decides everything downstream.

Curious whether that lines up with what other people are seeing.

[batchyard]

Can you fill in a bit more context? Are you testing this on 7900 XTX or something else?

[modrunner27]

I think the bottleneck is somewhere else. I've seen people blame tail latency when the real issue was much more mundane. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[rasterbench]

Can you fill in a bit more context? What model, what runtime, and roughly what context size?

[routecraft]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at speculative decoding. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[mangotinker73]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at speculative decoding. A lot o

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.

[amberpatch25]

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