Running my own instance of gpt-oss-20b on lmstudio thanks to nostr:nprofile1qqsgt0d4sa0pz0wvn9yck36xxezf3qhwzksu0q254vruqed5wvuavuspz3mhxue69uhhyetvv9ujuerpd46hxtnfduq3vamnwvaz7tmjv4kxz7fwwpexjmtpdshxuet5ut7hjf livestream “run your own ai”.
Need to see how far I can push this with bigger models.
Running an old ryzen 1600 with a 6700xt yields 40 tokens per second.
Super cool to run privately.