diff --git a/src/khoj/processor/conversation/openai/utils.py b/src/khoj/processor/conversation/openai/utils.py
index 84b0081b..82f68259 100644
--- a/src/khoj/processor/conversation/openai/utils.py
+++ b/src/khoj/processor/conversation/openai/utils.py
@@ -97,7 +97,8 @@ def completion_with_backoff(
     # Calculate cost of chat
     input_tokens = chunk.usage.prompt_tokens if hasattr(chunk, "usage") and chunk.usage else 0
     output_tokens = chunk.usage.completion_tokens if hasattr(chunk, "usage") and chunk.usage else 0
-    tracer["usage"] = get_chat_usage_metrics(model_name, input_tokens, output_tokens, tracer.get("usage"))
+    cost = chunk.usage.model_extra.get("estimated_cost") or 0  # Estimated costs returned by DeepInfra API
+    tracer["usage"] = get_chat_usage_metrics(model_name, input_tokens, output_tokens, tracer.get("usage"), cost)
 
     # Save conversation trace
     tracer["chat_model"] = model_name
@@ -208,7 +209,8 @@ def llm_thread(
         # Calculate cost of chat
         input_tokens = chunk.usage.prompt_tokens if hasattr(chunk, "usage") and chunk.usage else 0
         output_tokens = chunk.usage.completion_tokens if hasattr(chunk, "usage") and chunk.usage else 0
-        tracer["usage"] = get_chat_usage_metrics(model_name, input_tokens, output_tokens, tracer.get("usage"))
+        cost = chunk.usage.model_extra.get("estimated_cost") or 0  # Estimated costs returned by DeepInfra API
+        tracer["usage"] = get_chat_usage_metrics(model_name, input_tokens, output_tokens, tracer.get("usage"), cost)
 
         # Save conversation trace
         tracer["chat_model"] = model_name
diff --git a/src/khoj/utils/helpers.py b/src/khoj/utils/helpers.py
index 6fafa6d9..6214e5f5 100644
--- a/src/khoj/utils/helpers.py
+++ b/src/khoj/utils/helpers.py
@@ -584,13 +584,15 @@ def get_cost_of_chat_message(model_name: str, input_tokens: int = 0, output_toke
     return input_cost + output_cost + prev_cost
 
 
-def get_chat_usage_metrics(model_name: str, input_tokens: int = 0, output_tokens: int = 0, usage: dict = {}):
+def get_chat_usage_metrics(
+    model_name: str, input_tokens: int = 0, output_tokens: int = 0, usage: dict = {}, cost: float = None
+):
     """
-    Get usage metrics for chat message based on input and output tokens
+    Get usage metrics for chat message based on input and output tokens and cost
     """
     prev_usage = usage or {"input_tokens": 0, "output_tokens": 0, "cost": 0.0}
     return {
         "input_tokens": prev_usage["input_tokens"] + input_tokens,
         "output_tokens": prev_usage["output_tokens"] + output_tokens,
-        "cost": get_cost_of_chat_message(model_name, input_tokens, output_tokens, prev_cost=prev_usage["cost"]),
+        "cost": cost or get_cost_of_chat_message(model_name, input_tokens, output_tokens, prev_cost=prev_usage["cost"]),
     }