diff --git a/env.example b/env.example
index cd92abc8..955741ef 100644
--- a/env.example
+++ b/env.example
@@ -50,6 +50,7 @@
 # MAX_TOKEN_SUMMARY=500        # Max tokens for entity or relations summary
 # SUMMARY_LANGUAGE=English
 # MAX_EMBED_TOKENS=8192
+# ENABLE_LLM_CACHE_FOR_EXTRACT=false  # Enable LLM cache for entity extraction, defaults to false
 
 ### LLM Configuration (Use valid host. For local services installed with docker, you can use host.docker.internal)
 LLM_BINDING=ollama
diff --git a/lightrag/api/README.md b/lightrag/api/README.md
index f9a2a197..7a07ddb8 100644
--- a/lightrag/api/README.md
+++ b/lightrag/api/README.md
@@ -223,6 +223,11 @@ LightRAG supports binding to various LLM/Embedding backends:
 
 Use environment variables  `LLM_BINDING` or CLI argument `--llm-binding` to select LLM backend type. Use environment variables  `EMBEDDING_BINDING` or CLI argument `--embedding-binding` to select LLM backend type.
 
+### Entity Extraction Configuration
+* ENABLE_LLM_CACHE_FOR_EXTRACT: Enable LLM cache for entity extraction (default: false)
+
+It's very common to set `ENABLE_LLM_CACHE_FOR_EXTRACT` to true for test environment to reduce the cost of LLM calls.
+
 ### Storage Types Supported
 
 LightRAG uses 4 types of storage for difference purposes:
diff --git a/lightrag/api/lightrag_server.py b/lightrag/api/lightrag_server.py
index 2891b542..5df4f765 100644
--- a/lightrag/api/lightrag_server.py
+++ b/lightrag/api/lightrag_server.py
@@ -50,6 +50,9 @@ from .auth import auth_handler
 # This update allows the user to put a different.env file for each lightrag folder
 load_dotenv(".env", override=True)
 
+# Read entity extraction cache config
+enable_llm_cache = os.getenv("ENABLE_LLM_CACHE_FOR_EXTRACT", "false").lower() == "true"
+
 # Initialize config parser
 config = configparser.ConfigParser()
 config.read("config.ini")
@@ -323,7 +326,7 @@ def create_app(args):
             vector_db_storage_cls_kwargs={
                 "cosine_better_than_threshold": args.cosine_threshold
             },
-            enable_llm_cache_for_entity_extract=False,  # set to True for debuging to reduce llm fee
+            enable_llm_cache_for_entity_extract=enable_llm_cache,  # Read from environment variable
             embedding_cache_config={
                 "enabled": True,
                 "similarity_threshold": 0.95,
@@ -352,7 +355,7 @@ def create_app(args):
             vector_db_storage_cls_kwargs={
                 "cosine_better_than_threshold": args.cosine_threshold
             },
-            enable_llm_cache_for_entity_extract=False,  # set to True for debuging to reduce llm fee
+            enable_llm_cache_for_entity_extract=enable_llm_cache,  # Read from environment variable
             embedding_cache_config={
                 "enabled": True,
                 "similarity_threshold": 0.95,