multimodal local stack that feels real

[mangohouse]

Spent some time on this today and wanted to compare notes.

I'm comparing two approaches here: keep it simple on A6000, or squeeze more out of the stack and deal with the complexity.

Trying to avoid rebuilding the whole thing twice.

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

[visionforge]

That matches what I was seeing. I ended up in about the same place using llama.cpp. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[oxideworks59]

I'd start with the simpler stack.

[cachelab]

that fixed it for me too.

[mangohouse]

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

[turboarch95]

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

[latentpeak22]

What worked for me was pretty simple. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives.

[quiettoken]

following this because I'm in the same boat.

[mangohouse]

following this because I'm in the same boat.

Good point. I ended up in about the same place using LM Studio. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[thetaember38]

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