diff --git a/config/config.yaml b/config/config.yaml
index ef8575e43..8fd208c59 100644
--- a/config/config.yaml
+++ b/config/config.yaml
@@ -77,8 +77,8 @@ RPM: 10
 
 #### for Stable Diffusion
 ## Use SD service, based on https://github.com/AUTOMATIC1111/stable-diffusion-webui
-SD_URL: "YOUR_SD_URL"
-SD_T2I_API: "/sdapi/v1/txt2img"
+#SD_URL: "YOUR_SD_URL"
+#SD_T2I_API: "/sdapi/v1/txt2img"
 
 #### for Execution
 #LONG_TERM_MEMORY: false
@@ -93,8 +93,8 @@ SD_T2I_API: "/sdapi/v1/txt2img"
 # CALC_USAGE: false
 
 ### for Research
-MODEL_FOR_RESEARCHER_SUMMARY: gpt-3.5-turbo
-MODEL_FOR_RESEARCHER_REPORT: gpt-3.5-turbo-16k
+# MODEL_FOR_RESEARCHER_SUMMARY: gpt-3.5-turbo
+# MODEL_FOR_RESEARCHER_REPORT: gpt-3.5-turbo-16k
 
 ### choose the engine for mermaid conversion, 
 # default is nodejs, you can change it to playwright,pyppeteer or ink
@@ -108,4 +108,4 @@ MODEL_FOR_RESEARCHER_REPORT: gpt-3.5-turbo-16k
 ### repair operation on the content extracted from LLM's raw output. Warning, it improves the result but not fix all cases.
 # REPAIR_LLM_OUTPUT: false
 
-PROMPT_FORMAT: json #json or markdown
\ No newline at end of file
+# PROMPT_FORMAT: json #json or markdown
\ No newline at end of file
diff --git a/metagpt/config.py b/metagpt/config.py
index 8ad42c99f..19bd02c87 100644
--- a/metagpt/config.py
+++ b/metagpt/config.py
@@ -114,7 +114,7 @@ class Config(metaclass=Singleton):
         self.pyppeteer_executable_path = self._get("PYPPETEER_EXECUTABLE_PATH", "")
 
         self.repair_llm_output = self._get("REPAIR_LLM_OUTPUT", False)
-        self.prompt_format = self._get("PROMPT_FORMAT", "markdown")
+        self.prompt_format = self._get("PROMPT_FORMAT", "json")
         self.workspace_path = Path(self._get("WORKSPACE_PATH", DEFAULT_WORKSPACE_ROOT))
         self._ensure_workspace_exists()