VRAM bottleneck on consumer cards for Qwen

[torchshed]

My first guess would be this: Before rebuilding anything, I'd test one clean baseline with DeepSeek and compare from there. Not glamorous advice, but

Maybe, but I had the opposite result. I've seen people blame VRAM headroom when the real issue was much more mundane. Worth testing the simpler explanation before adopting the more interesting one.