Comparison
Base modelvsInstruct model / chat model
Base model
you load a model that never went through chat tuning, ask it a question, and it continues your sentence instead of answering it.
A model straight out of pretraining, before any chat tuning. It only continues text; ask it a question and it may reply with another question, because that is what raw internet text often does. You will rarely touch one directly.
Full entry →Instruct model / chat model
you never had to teach it how a conversation works; every model behind an API you call was already tuned to answer.
A base model that went through post-training to follow instructions, converse in turns and refuse harmful requests. Every model behind an API you call is one of these.
Full entry →