Fix OpenAI model instantiation in examples (#601)

dottxt-ai · Feb 1, 2024 · b5ab675 · b5ab675
1 parent c2b0f4f
commit b5ab675
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/docs/reference/choices.md b/docs/reference/choices.md
@@ -6,7 +6,7 @@ In some cases we know the output is to be chosen between different options. We c
 ```python
 import outlines.models as models
 
-complete = models.text_completion.openai("text-davinci-002")
+complete = models.openai("gpt-3.5-turbo")
 answer = complete(
     "Pick the odd word out: skirt, dress, pen, jacket",
     is_in=["skirt", "dress", "pen", "jacket"]

diff --git a/docs/reference/index.md b/docs/reference/index.md
@@ -10,6 +10,6 @@ By default, language models stop generating tokens after and <EOS> token was gen
 ```python
 import outlines.models as models
 
-complete = models.text_completion.openai("text-davinci-002")
+complete = models.openai("gpt-3.5-turbo")
 expert = complete("Name an expert in quantum gravity.", stop_at=["\n", "."])
 ```
diff --git a/docs/reference/types.md b/docs/reference/types.md
@@ -5,7 +5,7 @@ We can ask completions to be restricted to valid integers or floating-point numb
 ```python
 import outlines.models as models
 
-complete = models.text_completion.openai("text-davinci-002")
+complete = models.openai("gpt-3.5-turbo")
 answer = complete(
     "When I was 6 my sister was half my age. Now I’m 70 how old is my sister?",
     type="int"