diff --git a/llama_stack/templates/open-benchmark/open_benchmark.py b/llama_stack/templates/open-benchmark/open_benchmark.py
index 09ea972fa..b339e8c80 100644
--- a/llama_stack/templates/open-benchmark/open_benchmark.py
+++ b/llama_stack/templates/open-benchmark/open_benchmark.py
@@ -173,7 +173,7 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="huggingface",
             purpose=DatasetPurpose.eval_messages_answer,
             source=URIDataSource(
-                uri="huggingface://llamastack/simpleqa?split=train",
+                uri="huggingface://datasets/llamastack/simpleqa?split=train",
             ),
         ),
         DatasetInput(
@@ -181,7 +181,7 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="huggingface",
             purpose=DatasetPurpose.eval_messages_answer,
             source=URIDataSource(
-                uri="huggingface://llamastack/mmlu_cot?split=test&name=all",
+                uri="huggingface://datasets/llamastack/mmlu_cot?split=test&name=all",
             ),
         ),
         DatasetInput(
@@ -189,7 +189,7 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="huggingface",
             purpose=DatasetPurpose.eval_messages_answer,
             source=URIDataSource(
-                uri="huggingface://llamastack/gpqa_0shot_cot?split=test&name=gpqa_main",
+                uri="huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main",
             ),
         ),
         DatasetInput(
@@ -197,7 +197,7 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="huggingface",
             purpose=DatasetPurpose.eval_messages_answer,
             source=URIDataSource(
-                uri="huggingface://llamastack/math_500?split=test",
+                uri="huggingface://datasets/llamastack/math_500?split=test",
             ),
         ),
         DatasetInput(
@@ -205,7 +205,7 @@ def get_distribution_template() -> DistributionTemplate:
             provider_id="huggingface",
             purpose=DatasetPurpose.eval_messages_answer,
             source=URIDataSource(
-                uri="huggingface://llamastack/bfcl_v3?split=train",
+                uri="huggingface://datasets/llamastack/bfcl_v3?split=train",
             ),
         ),
     ]
diff --git a/llama_stack/templates/open-benchmark/run.yaml b/llama_stack/templates/open-benchmark/run.yaml
index 3af70d565..93f437273 100644
--- a/llama_stack/templates/open-benchmark/run.yaml
+++ b/llama_stack/templates/open-benchmark/run.yaml
@@ -161,35 +161,35 @@ datasets:
 - purpose: eval/messages-answer
   source:
     type: uri
-    uri: huggingface://llamastack/simpleqa?split=train
+    uri: huggingface://datasets/llamastack/simpleqa?split=train
   metadata: {}
   dataset_id: simpleqa
   provider_id: huggingface
 - purpose: eval/messages-answer
   source:
     type: uri
-    uri: huggingface://llamastack/mmlu_cot?split=test&name=all
+    uri: huggingface://datasets/llamastack/mmlu_cot?split=test&name=all
   metadata: {}
   dataset_id: mmlu_cot
   provider_id: huggingface
 - purpose: eval/messages-answer
   source:
     type: uri
-    uri: huggingface://llamastack/gpqa_0shot_cot?split=test&name=gpqa_main
+    uri: huggingface://datasets/llamastack/gpqa_0shot_cot?split=test&name=gpqa_main
   metadata: {}
   dataset_id: gpqa_cot
   provider_id: huggingface
 - purpose: eval/messages-answer
   source:
     type: uri
-    uri: huggingface://llamastack/math_500?split=test
+    uri: huggingface://datasets/llamastack/math_500?split=test
   metadata: {}
   dataset_id: math_500
   provider_id: huggingface
 - purpose: eval/messages-answer
   source:
     type: uri
-    uri: huggingface://llamastack/bfcl_v3?split=train
+    uri: huggingface://datasets/llamastack/bfcl_v3?split=train
   metadata: {}
   dataset_id: bfcl
   provider_id: huggingface