multimodal local stack that feels real for Qwen

[mangohouse]

Posting this in case it saves someone else a little time.

I'm not trying to build a perfect architecture diagram. I just want a sane starting point around TensorRT-LLM and GB200.

Would rather learn from your scars than collect my own all at once.

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

[cobaltworks25]

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

[phasespark]

worth checking the driver first.