---
type: learning
area: learning
status: learning
date: 2026-06-06
created: 2026-06-06
updated: 2026-06-06
tags:
  - learning
---
chat completion is used to simulate a conversation
user role is where the prompt should be written

to get token counts  for input -> reponse.usage.prompt_tokens 

! always estimate costs

parametters:
	- max_completion_tokens = controls length of response
	- temperature = ranges from 0 -> 2 (randomness)

Prompting Techniques
	-