It's unclear if the Recommended Usage section applies to the distilled models or not? also is the syntax for templating the same as the R1 or the base distilled model? do we use the llama3 and qwen2.5 templates or the deepseek-r1 templates?
distilled
R1
llama3
qwen2.5
deepseek-r1
· Sign up or log in to comment
