Add another edge case

deepset-ai · Mar 7, 2024 · fed8475 · fed8475
1 parent 336e0b7
commit fed8475
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 2 deletions.
diff --git a/haystack/utils/openai_utils.py b/haystack/utils/openai_utils.py
@@ -89,6 +89,8 @@ def _openai_text_completion_tokenization_details(model_name: str):
             or model_name == "gpt-35-turbo-0613"
         ):
             max_tokens_limit = 4096
+        if model_name == "gpt-3.5-turbo-16k" or model_name == "gpt-35-turbo-16k":
+            max_tokens_limit = 16384
 
         # GPT-4 models that have a different token limit than 4096
         # Ref: https://platform.openai.com/docs/models/gpt-4

diff --git a/test/utils/test_openai_utils.py b/test/utils/test_openai_utils.py
@@ -24,6 +24,7 @@
         ("gpt-3.5-turbo-0613", "cl100k_base", 4096),
         ("gpt-3.5-turbo", "cl100k_base", 16385),
         ("gpt-3.5-turbo-1106", "cl100k_base", 16385),
+        ("gpt-3.5-turbo-16k", "cl100k_base", 16384),
         # GPT 4
         ("gpt-4-32k", "cl100k_base", 32768),
         ("gpt-4-1106", "cl100k_base", 128000),
@@ -32,9 +33,8 @@
         ("gpt-4", "cl100k_base", 8192),
         # GPT-35 Azure
         ("gpt-35-turbo-instruct", "cl100k_base", 4096),
-        ("gpt-35-turbo-0613", "cl100k_base", 4096),
         ("gpt-35-turbo", "cl100k_base", 16385),
-        ("gpt-35-turbo-1106", "cl100k_base", 16385),
+        ("gpt-35-turbo-16k", "cl100k_base", 16384),
         # davinci and babbage
         ("davinci-002", "cl100k_base", 16384),
         ("babbage-002", "cl100k_base", 16384),