Improve redis semantic cache implementation

.circleci/requirements.txt

@@ -5,6 +5,7 @@ tiktoken
 importlib_metadata
 cohere
 redis
+redisvl==0.3.3
 anthropic
 orjson==3.9.15
 pydantic==2.7.1

Dockerfile

@@ -35,9 +35,6 @@ RUN pip install dist/*.whl
 # install dependencies as wheels
 RUN pip wheel --no-cache-dir --wheel-dir=/wheels/ -r requirements.txt
 
-# install semantic-cache [Experimental]- we need this here and not in requirements.txt because redisvl pins to pydantic 1.0 
-RUN pip install redisvl==0.0.7 --no-deps
-
 # ensure pyjwt is used, not jwt
 RUN pip uninstall jwt -y
 RUN pip uninstall PyJWT -y

Dockerfile.database

@@ -50,9 +50,6 @@ COPY --from=builder /wheels/ /wheels/
 # Install the built wheel using pip; again using a wildcard if it's the only file
 RUN pip install *.whl /wheels/* --no-index --find-links=/wheels/ && rm -f *.whl && rm -rf /wheels
 
-# install semantic-cache [Experimental]- we need this here and not in requirements.txt because redisvl pins to pydantic 1.0 
-RUN pip install redisvl==0.0.7 --no-deps
-
 # ensure pyjwt is used, not jwt
 RUN pip uninstall jwt -y
 RUN pip uninstall PyJWT -y

Dockerfile.non_root

@@ -50,9 +50,6 @@ COPY --from=builder /wheels/ /wheels/
 # Install the built wheel using pip; again using a wildcard if it's the only file
 RUN pip install *.whl /wheels/* --no-index --find-links=/wheels/ && rm -f *.whl && rm -rf /wheels
 
-# install semantic-cache [Experimental]- we need this here and not in requirements.txt because redisvl pins to pydantic 1.0 
-RUN pip install redisvl==0.0.7 --no-deps
-
 # ensure pyjwt is used, not jwt
 RUN pip uninstall jwt -y
 RUN pip uninstall PyJWT -y

docs/my-website/docs/caching/all_caches.md

@@ -30,15 +30,21 @@ import litellm
 from litellm import completion
 from litellm.caching import Cache
 
-litellm.cache = Cache(type="redis", host=<host>, port=<port>, password=<password>)
+litellm.cache = Cache(
+    type="redis",
+    host=os.environ["REDIS_HOST"],
+    port=os.environ["REDIS_PORT"],
+    password=os.environ["REDIS_PASSWORD"],
+    ttl=60
+)
 
 # Make completion calls
 response1 = completion(
-    model="gpt-3.5-turbo", 
+    model="gpt-3.5-turbo",
     messages=[{"role": "user", "content": "Tell me a joke."}]
 )
 response2 = completion(
-    model="gpt-3.5-turbo", 
+    model="gpt-3.5-turbo",
     messages=[{"role": "user", "content": "Tell me a joke."}]
 )
 
@@ -72,11 +78,11 @@ litellm.cache = Cache(type="s3", s3_bucket_name="cache-bucket-litellm", s3_regio
 
 # Make completion calls
 response1 = completion(
-    model="gpt-3.5-turbo", 
+    model="gpt-3.5-turbo",
     messages=[{"role": "user", "content": "Tell me a joke."}]
 )
 response2 = completion(
-    model="gpt-3.5-turbo", 
+    model="gpt-3.5-turbo",
     messages=[{"role": "user", "content": "Tell me a joke."}]
 )
 
@@ -90,10 +96,10 @@ response2 = completion(
 
 Install redis
 ```shell
-pip install redisvl==0.0.7
+pip install redisvl==0.3.3
 ```
 
-For the hosted version you can setup your own Redis DB here: https://app.redislabs.com/
+For the hosted version you can setup your own Redis DB here: https://redis.io/try-free/
 
 ```python
 import litellm
@@ -110,6 +116,7 @@ litellm.cache = Cache(
     host=os.environ["REDIS_HOST"],
     port=os.environ["REDIS_PORT"],
     password=os.environ["REDIS_PASSWORD"],
+    ttl=60,
     similarity_threshold=0.8, # similarity threshold for cache hits, 0 == no similarity, 1 = exact matches, 0.5 == 50% similarity
     redis_semantic_cache_embedding_model="text-embedding-ada-002", # this model is passed to litellm.embedding(), any litellm.embedding() model is supported here
 )

litellm/_redis.py

@@ -120,11 +120,14 @@ def _get_redis_client_logic(**env_overrides):
         **env_overrides,
     }
 
-    if "url" in redis_kwargs and redis_kwargs["url"] is not None:
-        redis_kwargs.pop("host", None)
-        redis_kwargs.pop("port", None)
-        redis_kwargs.pop("db", None)
-        redis_kwargs.pop("password", None)
+    if "url" in redis_kwargs:
+        if redis_kwargs["url"] is not None:
+            redis_kwargs.pop("host", None)
+            redis_kwargs.pop("port", None)
+            redis_kwargs.pop("db", None)
+            redis_kwargs.pop("password", None)
+        else:
+            redis_kwargs.pop("url", None)
     elif "host" not in redis_kwargs or redis_kwargs["host"] is None:
         raise ValueError("Either 'host' or 'url' must be specified for redis.")
     # litellm.print_verbose(f"redis_kwargs: {redis_kwargs}")
@@ -157,6 +160,7 @@ def get_redis_client(**env_overrides):
             new_startup_nodes.append(ClusterNode(**item))
         redis_kwargs.pop("startup_nodes")
         return redis.RedisCluster(startup_nodes=new_startup_nodes, **cluster_kwargs)
+
     return redis.Redis(**redis_kwargs)