privacy wins of local models: what are you seeing?

[rasterbench]

Not sure I agree there. The reported numbers can look compelling, but in ordinary usage I didn't see the same benefit. Not saying you're wrong, just t

What exact setup are you running? Are you testing this on A6000 or something else?

[cachebench]

that fixed it for me too.

[rocmhouse90]

My first guess would be this: Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at MoE serving. Once you know the plain path works, then add the faster path back in piece by piece.

[audioriver]

A couple things I'd check first. If you're on TensorRT-LLM, 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.

[mangoquill]

A couple things I'd check first. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. A l

Appreciate the detail. I ended up in about the same place using SGLang. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[rasterbench]

What worked for me was pretty simple. If you're on llama.cpp, I'd confirm the version first and then reduce variables until the behavior is obvious.