You've reached the end!
Saw it on here a few days ago, you put in your GPUs and it helps give you an idea of what you can run. Although, the specific model weight is dramatically altered by use case. Do you want code creation? Creative writing? Assistant workloads? I would suggest doing some Googles/AI searches on “[Model returned by RunItLocal] for [use case]”.
spottedmarley•2h ago
donkan168•2h ago
spottedmarley•1h ago
Qwen4 should be dropping soon, hopefully another 27b that is less aggressively eager than 3.8-flash-next
donkan168•1h ago
spottedmarley•45m ago
donkan168•27m ago