You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Google Gemini's aggressive rate limits (forcing >60s waits per call) have made real-time API dependency
692
-
unsustainable.
693
-
I am now shifting focus to <strong>Supervised Fine-Tuning (SFT)</strong> a local model to handle categorization
694
-
completely offline, eliminating rate limits and latency forever.
691
+
Google Gemini's increased rate limits :( Even switching to an RL agent does not benifit much. Now shifting on SFT <strong>unsloth/Meta-Llama-3.1-8B-Instruct-bnb-4bit.</strong>
0 commit comments