framework overhead in real deployments

[llm_wren]

Did a bit of testing and thought I'd post the result here.

After a few runs on dual 3090, I ended up with decent performance, but the setup only really made sense once I adjusted around tokens/sec.

Happy to post more details if anyone wants them.

My use case is pretty ordinary: local chat, some document work, and occasional batch jobs. I keep promising myself this is the last rebuild.

[cacheworks]

worth checking the driver first.

[cobaltworks25]

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

There are a few different conversations hiding inside this thread, so being explicit about the goal helps.

[rocmhouse90]

I'd approach it in a boring order. Start by verifying the driver/runtime combo, then test with a smaller prompt, then look at tokens/sec. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

If you post your exact settings, people here can probably narrow it down pretty fast.

[mangoquill]

What worked for me was pretty simple. Before rebuilding anything, I'd test one clean baseline with Phi and compare from there. A lot of these threads end up being one bad config plus a pile of conclusions drawn from it.

[vramforge]

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.