Despite the bullshit excuses, it is clear for all to see that the power of llms is being increasingly censored, perhaps for good reason.
I don't want to speculate about the future, but I suggest making local copies of the most powerful and efficient open source models currently available. TAKE THE WEIGHTS.
Now as long as you can access compute you can access inference.
Resources:
Comparison: https://artificialanalysis.ai/
Chat Interface: https://www.librechat.ai/
Signature
Paper hands die young