fix: max_tokens in O1 (#17703)

run-llama · Feb 3, 2025 · 8a3feea · 8a3feea
1 parent 4c81fff
commit 8a3feea
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/llama-index-integrations/llms/llama-index-llms-openai/llama_index/llms/openai/base.py b/llama-index-integrations/llms/llama-index-llms-openai/llama_index/llms/openai/base.py
@@ -424,7 +424,7 @@ def _get_model_kwargs(self, **kwargs: Any) -> Dict[str, Any]:
         all_kwargs = {**base_kwargs, **self.additional_kwargs}
         if "stream" not in all_kwargs and "stream_options" in all_kwargs:
             del all_kwargs["stream_options"]
-        if self.model in O1_MODELS and base_kwargs["max_tokens"] is not None:
+        if self.model in O1_MODELS and base_kwargs.get("max_tokens") is not None:
             # O1 models use max_completion_tokens instead of max_tokens
             all_kwargs["max_completion_tokens"] = all_kwargs.get(
                 "max_completion_tokens", all_kwargs["max_tokens"]

diff --git a/llama-index-integrations/llms/llama-index-llms-openai/pyproject.toml b/llama-index-integrations/llms/llama-index-llms-openai/pyproject.toml
@@ -29,7 +29,7 @@ exclude = ["**/BUILD"]
 license = "MIT"
 name = "llama-index-llms-openai"
 readme = "README.md"
-version = "0.3.16"
+version = "0.3.17"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"