prompt injection in local agent setups

[cacheworks]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at speculative decoding. Once you know the plain path works, then add the faster path back in piece by piece. Turns out 'just one more test' was a lie.

[fastquill74]

Tried to reproduce it just now. On mini PC with SGLang, I saw basically the same behavior. The surprise was that memory pressure showed up before throughput collapsed.

I'm happy to test one more variant if you want a second data point.

[rocmriley]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at speculative decoding. On

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

[cudaline]

A couple things I'd check first. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at consumer GPUs.

[cozygrid12]

Interesting, I hadn't tried that. The pattern around tail latency has been pretty similar for me. Worth rechecking your defaults though, because one bad assumption can make the rest of the thread look confusing.

[latentpeak22]

My first guess would be this: On dual 3090, the easiest win is often to back off the aggressive settings and see whether the problem survives.