diff --git a/agenta-backend/agenta_backend/models/api/evaluation_model.py b/agenta-backend/agenta_backend/models/api/evaluation_model.py
index 4549e85842..2016d49bc2 100644
--- a/agenta-backend/agenta_backend/models/api/evaluation_model.py
+++ b/agenta-backend/agenta_backend/models/api/evaluation_model.py
@@ -72,6 +72,7 @@ class Evaluation(BaseModel):
     status: Result
     aggregated_results: List[AggregatedResult]
     average_cost: Optional[Result]
+    total_cost: Optional[Result]
     average_latency: Optional[Result]
     created_at: datetime
     updated_at: datetime
diff --git a/agenta-backend/agenta_backend/models/converters.py b/agenta-backend/agenta_backend/models/converters.py
index 2e13547ff5..541462c3b0 100644
--- a/agenta-backend/agenta_backend/models/converters.py
+++ b/agenta-backend/agenta_backend/models/converters.py
@@ -141,6 +141,7 @@ async def evaluation_db_to_pydantic(
         created_at=evaluation_db.created_at,
         updated_at=evaluation_db.updated_at,
         average_cost=evaluation_db.average_cost,
+        total_cost=evaluation_db.total_cost,
         average_latency=evaluation_db.average_latency,
     )
 
diff --git a/agenta-backend/agenta_backend/models/db_models.py b/agenta-backend/agenta_backend/models/db_models.py
index f7dcc49ba9..e67c8b42ce 100644
--- a/agenta-backend/agenta_backend/models/db_models.py
+++ b/agenta-backend/agenta_backend/models/db_models.py
@@ -269,6 +269,7 @@ class EvaluationDB(Document):
     evaluators_configs: List[PydanticObjectId]
     aggregated_results: List[AggregatedResult]
     average_cost: Optional[Result] = None
+    total_cost: Optional[Result] = None
     average_latency: Optional[Result] = None
     created_at: datetime = Field(default=datetime.now(timezone.utc))
     updated_at: datetime = Field(default=datetime.now(timezone.utc))
diff --git a/agenta-backend/agenta_backend/services/aggregation_service.py b/agenta-backend/agenta_backend/services/aggregation_service.py
index b459766301..d1120f1b53 100644
--- a/agenta-backend/agenta_backend/services/aggregation_service.py
+++ b/agenta-backend/agenta_backend/services/aggregation_service.py
@@ -99,3 +99,29 @@ def aggregate_float_from_llm_app_response(
             value=None,
             error=Error(message=str(exc), stacktrace=str(traceback.format_exc())),
         )
+
+
+def sum_float_from_llm_app_response(
+    invocation_results: List[InvokationResult], key: Optional[str]
+) -> Result:
+    try:
+        if not key:
+            raise ValueError("Key is required to aggregate InvokationResult objects.")
+
+        values = [
+            getattr(inv_result, key)
+            for inv_result in invocation_results
+            if hasattr(inv_result, key) and getattr(inv_result, key) is not None
+        ]
+
+        if not values:
+            raise ValueError(f"No valid values found for {key} sum aggregation.")
+
+        total_value = sum(values)
+        return Result(type=key, value=total_value)
+    except Exception as exc:
+        return Result(
+            type="error",
+            value=None,
+            error=Error(message=str(exc), stacktrace=str(traceback.format_exc())),
+        )
diff --git a/agenta-backend/agenta_backend/tasks/evaluations.py b/agenta-backend/agenta_backend/tasks/evaluations.py
index 956a4276d0..5dd602de6e 100644
--- a/agenta-backend/agenta_backend/tasks/evaluations.py
+++ b/agenta-backend/agenta_backend/tasks/evaluations.py
@@ -285,10 +285,17 @@ def evaluate(
         average_cost = aggregation_service.aggregate_float_from_llm_app_response(
             app_outputs, "cost"
         )
+        total_cost = aggregation_service.sum_float_from_llm_app_response(
+            app_outputs, "cost"
+        )
         loop.run_until_complete(
             update_evaluation(
                 evaluation_id,
-                {"average_latency": average_latency, "average_cost": average_cost},
+                {
+                    "average_latency": average_latency,
+                    "average_cost": average_cost,
+                    "total_cost": total_cost,
+                },
             )
         )
 
diff --git a/agenta-cli/agenta/client/backend/types/evaluation.py b/agenta-cli/agenta/client/backend/types/evaluation.py
index ca877a97f2..55bb6685cb 100644
--- a/agenta-cli/agenta/client/backend/types/evaluation.py
+++ b/agenta-cli/agenta/client/backend/types/evaluation.py
@@ -27,6 +27,7 @@ class Evaluation(pydantic.BaseModel):
     status: Result
     aggregated_results: typing.List[AggregatedResult]
     average_cost: typing.Optional[Result]
+    total_cost: typing.Optional[Result]
     average_latency: typing.Optional[Result]
     created_at: dt.datetime
     updated_at: dt.datetime
diff --git a/agenta-web/src/components/pages/evaluations/evaluationResults/EvaluationResults.tsx b/agenta-web/src/components/pages/evaluations/evaluationResults/EvaluationResults.tsx
index 8ea9dc84db..bfb7965239 100644
--- a/agenta-web/src/components/pages/evaluations/evaluationResults/EvaluationResults.tsx
+++ b/agenta-web/src/components/pages/evaluations/evaluationResults/EvaluationResults.tsx
@@ -289,7 +289,7 @@ const EvaluationResults: React.FC<Props> = () => {
             {
                 flex: 1,
                 field: "average_latency",
-                headerName: "Latency",
+                headerName: "Avg. Latency",
                 hide: hiddenCols.includes("Latency"),
                 minWidth: 120,
                 ...getFilterParams("number"),
@@ -297,12 +297,12 @@ const EvaluationResults: React.FC<Props> = () => {
             },
             {
                 flex: 1,
-                field: "average_cost",
-                headerName: "Cost",
+                field: "total_cost",
+                headerName: "Total Cost",
                 hide: hiddenCols.includes("Cost"),
                 minWidth: 120,
                 ...getFilterParams("number"),
-                valueGetter: (params) => getTypedValue(params?.data?.average_cost),
+                valueGetter: (params) => getTypedValue(params?.data?.total_cost),
             },
             {
                 flex: 1,
diff --git a/agenta-web/src/lib/Types.ts b/agenta-web/src/lib/Types.ts
index 8a981fa524..a5d9bb95f1 100644
--- a/agenta-web/src/lib/Types.ts
+++ b/agenta-web/src/lib/Types.ts
@@ -409,6 +409,7 @@ export interface _Evaluation {
     revisions: string[]
     average_latency?: TypedValue & {error: null | EvaluationError}
     average_cost?: TypedValue & {error: null | EvaluationError}
+    total_cost?: TypedValue & {error: null | EvaluationError}
     variant_revision_ids: string[]
 }
 
diff --git a/agenta-web/src/services/evaluations/index.ts b/agenta-web/src/services/evaluations/index.ts
index 7a2089544f..c672840052 100644
--- a/agenta-web/src/services/evaluations/index.ts
+++ b/agenta-web/src/services/evaluations/index.ts
@@ -105,6 +105,7 @@ const evaluationTransformer = (item: any) => ({
     variant_revision_ids: item.variant_revision_ids,
     variant_ids: item.variant_ids,
     average_cost: item.average_cost,
+    total_cost: item.total_cost,
     average_latency: item.average_latency,
 })
 export const fetchAllEvaluations = async (appId: string) => {