I'd have to say I'm very impressed with WizardLM 30B (the newer one). I run it in GPT4ALL, and even though it is slow the results are quite impressive. Looking forward to Orca 13b if it ever releases!
Wonder what steam deck support will look like.