do you even need a vector DB here on B200

[sanepeak48]

Maybe, but I had the opposite result. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Not saying you're wrong, just that I wouldn't lock that in as the lesson yet.

[southtinker]

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 NVFP4 decides everything downstream.

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

I care more about stability than squeezing the last possible tokens/sec out of it.

[rasterhall]

Can you fill in a bit more context? Are you testing this on 5090 or something else?

[batchscope72]

That matches what I was seeing. That matches what I saw on 5090.

[thetafield]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Kimi and compare from there. Not glamorous advice, but repeatable baselines save a lot of time.

[llm_wren]

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

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

[papermixer4]

What worked for me was pretty simple. On 5090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[thetanest]

that fixed it for me too.

[southspark]

Can you fill in a bit more context? Is this local-only or are you routing anything externally?

[paperport64]

That matches what I was seeing. That matches what I saw on 5090.

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.