Context

#1
by saishf - opened

1.png
Does the 32K context in this image apply to the llama-3 model?

no, there's plans to train qwen 72B

8k for llama-3, 32k for qwen.

Sign up or log in to comment