openai[patch]: fix azure embedding length check (#19870)

libs/partners/openai/langchain_openai/embeddings/azure.py

@@ -1,4 +1,5 @@
 """Azure OpenAI embeddings wrapper."""
+
 from __future__ import annotations
 
 import os
@@ -57,6 +58,8 @@ class AzureOpenAIEmbeddings(OpenAIEmbeddings):
     openai_api_version: Optional[str] = Field(default=None, alias="api_version")
     """Automatically inferred from env var `OPENAI_API_VERSION` if not provided."""
     validate_base_url: bool = True
+    chunk_size: int = 2048
+    """Maximum number of texts to embed in each batch"""
 
     @root_validator()
     def validate_environment(cls, values: Dict) -> Dict:
@@ -102,7 +105,11 @@ class AzureOpenAIEmbeddings(OpenAIEmbeddings):
         # Azure OpenAI embedding models allow a maximum of 2048 texts
         # at a time in each batch
         # See: https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/embeddings?tabs=console#best-practices
-        values["chunk_size"] = min(values["chunk_size"], 2048)
+        if values["chunk_size"] > 2048:
+            raise ValueError(
+                "Azure OpenAI embeddings only allow a maximum of 2048 texts at a time "
+                "in each batch."
+            )
         # For backwards compatibility. Before openai v1, no distinction was made
         # between azure_endpoint and base_url (openai_api_base).
         openai_api_base = values["openai_api_base"]
@@ -126,12 +133,16 @@ class AzureOpenAIEmbeddings(OpenAIEmbeddings):
             "api_version": values["openai_api_version"],
             "azure_endpoint": values["azure_endpoint"],
             "azure_deployment": values["deployment"],
-            "api_key": values["openai_api_key"].get_secret_value()
-            if values["openai_api_key"]
-            else None,
-            "azure_ad_token": values["azure_ad_token"].get_secret_value()
-            if values["azure_ad_token"]
-            else None,
+            "api_key": (
+                values["openai_api_key"].get_secret_value()
+                if values["openai_api_key"]
+                else None
+            ),
+            "azure_ad_token": (
+                values["azure_ad_token"].get_secret_value()
+                if values["azure_ad_token"]
+                else None
+            ),
             "azure_ad_token_provider": values["azure_ad_token_provider"],
             "organization": values["openai_organization"],
             "base_url": values["openai_api_base"],

libs/partners/openai/tests/integration_tests/embeddings/test_azure.py

@@ -60,8 +60,8 @@ def test_azure_openai_embedding_documents_chunk_size() -> None:
     embedding = _get_embeddings()
     embedding.embedding_ctx_length = 8191
     output = embedding.embed_documents(documents)
-    # Max 16 chunks per batch on Azure OpenAI embeddings
-    assert embedding.chunk_size == 16
+    # Max 2048 chunks per batch on Azure OpenAI embeddings
+    assert embedding.chunk_size == 2048
     assert len(output) == 20
     assert all([len(out) == 1536 for out in output])