diff --git a/python/third_party/TradingAgents/.env.example b/python/third_party/TradingAgents/.env.example
index e921f45b1..2e9481546 100644
--- a/python/third_party/TradingAgents/.env.example
+++ b/python/third_party/TradingAgents/.env.example
@@ -19,11 +19,11 @@ FINNHUB_API_KEY=your_finnhub_api_key_here
 # because they don't support OpenAI's embeddings API
 # If not set, the system will try to use OPENAI_API_KEY for embeddings
 # Get your key from: https://platform.openai.com/api-keys
-EMBEDDINGS_API_KEY=your_openai_embeddings_key_here
+EMBEDDER_API_KEY=your_openai_embeddings_key_here
 # Default embedding URL is OpenAI
-EMBEDDINGS_BACKEND_URL=your_embedding_model_backend_url_here
+EMBEDDER_BASE_URL=your_embedding_model_backend_url_here
 # Default embedding model
-EMBEDDINGS_MODEL=text-embedding-3-small
+EMBEDDER_MODEL_ID=text-embedding-3-small
 
 
 # =============================================================================
@@ -118,7 +118,7 @@ TRADINGAGENTS_ONLINE_TOOLS=true
 # OpenRouter does not support OpenAI's embeddings API, which is required for
 # the memory functionality in TradingAgents. You MUST set both:
 # OPENAI_API_KEY=sk-or-v1-your-openrouter-key-here  (for chat models)
-# EMBEDDINGS_API_KEY=sk-your-real-openai-key-here  (for embeddings)
+# EMBEDDER_API_KEY=sk-your-real-openai-key-here  (for embeddings)
 
 # For Azure OpenAI
 # TRADINGAGENTS_LLM_PROVIDER=azure
@@ -136,7 +136,7 @@ TRADINGAGENTS_ONLINE_TOOLS=true
 
 # Solution 1 - Use dedicated OpenAI key for embeddings (RECOMMENDED):
 # OPENAI_API_KEY=sk-or-v1-your-openrouter-key-here
-# EMBEDDINGS_API_KEY=sk-your-real-openai-key-here
+# EMBEDDER_API_KEY=sk-your-real-openai-key-here
 
 # Solution 2 - Switch to Ollama for fully local processing:
 # TRADINGAGENTS_LLM_PROVIDER=ollama
@@ -144,7 +144,7 @@ TRADINGAGENTS_ONLINE_TOOLS=true
 # Then run: ollama pull llama3.1 && ollama pull nomic-embed-text
 
 # Solution 3 - Accept limited memory functionality:
-# If you don't set EMBEDDINGS_API_KEY, the system will use dummy
+# If you don't set EMBEDDER_API_KEY, the system will use dummy
 # embeddings and continue running, but memory features will be limited.
 
 # Affected providers that need separate embeddings key:
@@ -162,6 +162,6 @@ TRADINGAGENTS_ONLINE_TOOLS=true
 # 1. Copy this file to .env: cp .env.example .env
 # 2. Fill in your actual API keys and configuration values
 # 3. Make sure .env is in your .gitignore to keep your keys private
-# 4. For OpenRouter users: Set both OPENAI_API_KEY and EMBEDDINGS_API_KEY
+# 4. For OpenRouter users: Set both OPENAI_API_KEY and EMBEDDER_API_KEY
 # 5. For Ollama users: Run 'ollama pull nomic-embed-text' for embeddings support
 # 6. Run the application: python main.py or python -m cli.main
diff --git a/python/third_party/TradingAgents/tradingagents/agents/utils/memory.py b/python/third_party/TradingAgents/tradingagents/agents/utils/memory.py
index c5ab19be4..8ac117140 100644
--- a/python/third_party/TradingAgents/tradingagents/agents/utils/memory.py
+++ b/python/third_party/TradingAgents/tradingagents/agents/utils/memory.py
@@ -6,11 +6,11 @@
 
 class FinancialSituationMemory:
     def __init__(self, name, config):
-        self.embedding = config["embeddings_model"]
-        self.client = OpenAI(base_url=config["embeddings_backend_url"])
-        if "localhost" not in config["embeddings_backend_url"]:
-            embeddings_api_key = os.getenv("EMBEDDINGS_API_KEY") or os.getenv("OPENAI_API_KEY")
-            self.client = OpenAI(api_key=embeddings_api_key, base_url=config["embeddings_backend_url"])
+        self.embedding = config["EMBEDDER_MODEL_ID"]
+        self.client = OpenAI(base_url=config["EMBEDDER_BASE_URL"])
+        if "localhost" not in config["EMBEDDER_BASE_URL"]:
+            EMBEDDER_API_KEY = os.getenv("EMBEDDER_API_KEY") or os.getenv("OPENAI_API_KEY")
+            self.client = OpenAI(api_key=EMBEDDER_API_KEY, base_url=config["EMBEDDER_BASE_URL"])
 
         self.chroma_client = chromadb.Client(Settings(allow_reset=True))
         try:
diff --git a/python/third_party/TradingAgents/tradingagents/default_config.py b/python/third_party/TradingAgents/tradingagents/default_config.py
index 287a744a2..355d6899c 100644
--- a/python/third_party/TradingAgents/tradingagents/default_config.py
+++ b/python/third_party/TradingAgents/tradingagents/default_config.py
@@ -38,8 +38,8 @@ def str_to_bool(value):
     "quick_think_llm": os.getenv("TRADINGAGENTS_QUICK_THINK_LLM", "gpt-4o-mini"),
     "backend_url": os.getenv("TRADINGAGENTS_BACKEND_URL", "https://api.openai.com/v1"),
     # Embeddings settings
-    "embeddings_backend_url": os.getenv("EMBEDDINGS_BACKEND_URL", "https://api.openai.com/v1"),
-    "embeddings_model": os.getenv("EMBEDDINGS_MODEL", "text-embedding-3-small"),
+    "EMBEDDER_BASE_URL": os.getenv("EMBEDDER_BASE_URL", "https://api.openai.com/v1"),
+    "EMBEDDER_MODEL_ID": os.getenv("EMBEDDER_MODEL_ID", "text-embedding-3-small"),
     # Debate and discussion settings
     "max_debate_rounds": int(os.getenv("TRADINGAGENTS_MAX_DEBATE_ROUNDS", "1")),
     "max_risk_discuss_rounds": int(os.getenv("TRADINGAGENTS_MAX_RISK_DISCUSS_ROUNDS", "1")),