From 0c26404cb3ce8915d1baae9812ed993c2462c10f Mon Sep 17 00:00:00 2001
From: Ishaan Jaff <ishaanjaffer0324@gmail.com>
Date: Wed, 10 Apr 2024 16:44:40 -0700
Subject: [PATCH 1/2] fix - allow base64 cache hits embedding responses

---
 litellm/utils.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/litellm/utils.py b/litellm/utils.py
index eee7f8202e9..7ce98174cf9 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -236,7 +236,7 @@ class HiddenParams(OpenAIObject):
 
     class Config:
         extra = "allow"
-        protected_namespaces = ()      
+        protected_namespaces = ()
 
     def get(self, key, default=None):
         # Custom .get() method to access attributes with a default value if the attribute doesn't exist
@@ -606,7 +606,7 @@ class ModelResponse(OpenAIObject):
 
 
 class Embedding(OpenAIObject):
-    embedding: list = []
+    embedding: Union[list, str] = []
     index: int
     object: str
 

From 8bc02b34c2cea40c5e3418fb94f380489744a09d Mon Sep 17 00:00:00 2001
From: Ishaan Jaff <ishaanjaffer0324@gmail.com>
Date: Wed, 10 Apr 2024 16:46:56 -0700
Subject: [PATCH 2/2] test -base64 cache hits

---
 .../tests/test_openai_simple_embedding.py     | 10 ++++++
 litellm/tests/test_caching.py                 | 35 +++++++++++++++++++
 2 files changed, 45 insertions(+)
 create mode 100644 litellm/proxy/tests/test_openai_simple_embedding.py

diff --git a/litellm/proxy/tests/test_openai_simple_embedding.py b/litellm/proxy/tests/test_openai_simple_embedding.py
new file mode 100644
index 00000000000..7dd38c0b396
--- /dev/null
+++ b/litellm/proxy/tests/test_openai_simple_embedding.py
@@ -0,0 +1,10 @@
+import openai
+
+client = openai.OpenAI(api_key="sk-1234", base_url="http://0.0.0.0:4000")
+
+# # request sent to model set on litellm proxy, `litellm --model`
+response = client.embeddings.create(
+    model="text-embedding-ada-002", input=["test"], encoding_format="base64"
+)
+
+print(response)
diff --git a/litellm/tests/test_caching.py b/litellm/tests/test_caching.py
index 835d3611b83..2c3c863de5f 100644
--- a/litellm/tests/test_caching.py
+++ b/litellm/tests/test_caching.py
@@ -387,6 +387,41 @@ async def test_embedding_caching_azure_individual_items_reordered():
     assert embedding_val_1.data[1]["index"] == 1
 
 
+@pytest.mark.asyncio
+async def test_embedding_caching_base_64():
+    """ """
+    litellm.cache = Cache(
+        type="redis",
+        host=os.environ["REDIS_HOST"],
+        port=os.environ["REDIS_PORT"],
+    )
+    import uuid
+
+    inputs = [
+        f"{uuid.uuid4()} hello this is ishaan",
+        f"{uuid.uuid4()} hello this is ishaan again",
+    ]
+
+    embedding_val_1 = await aembedding(
+        model="azure/azure-embedding-model",
+        input=inputs,
+        caching=True,
+        encoding_format="base64",
+    )
+    embedding_val_2 = await aembedding(
+        model="azure/azure-embedding-model",
+        input=inputs,
+        caching=True,
+        encoding_format="base64",
+    )
+
+    assert embedding_val_2._hidden_params["cache_hit"] == True
+    print(embedding_val_2)
+    print(embedding_val_1)
+    assert embedding_val_2.data[0]["embedding"] == embedding_val_1.data[0]["embedding"]
+    assert embedding_val_2.data[1]["embedding"] == embedding_val_1.data[1]["embedding"]
+
+
 @pytest.mark.asyncio
 async def test_redis_cache_basic():
     """