Comment on Mistral Remove "Committing to open models" from their websiteparentComments−vitorgrs2yReally depends how good llama 3 will be. I see people thinking it would be better than GPT 4. But they still need to build a model as good as GPT 3.5 even, as llama 2 is... not.
Comments
Really depends how good llama 3 will be. I see people thinking it would be better than GPT 4. But they still need to build a model as good as GPT 3.5 even, as llama 2 is... not.