diff --git a/docs/sample_recipes/vllm-open-access-hf-model-medium b/docs/sample_recipes/vllm-open-access-hf-model-medium
new file mode 100644
index 0000000..f2e35e0
--- /dev/null
+++ b/docs/sample_recipes/vllm-open-access-hf-model-medium
@@ -0,0 +1,30 @@
+{
+    "recipe_id": "llm_inference_nvidia",
+    "recipe_mode": "service",
+    "deployment_name": "vllm-from-hf-medium",
+    "recipe_image_uri": "iad.ocir.io/iduyx1qnmway/corrino-devops-repository:vllmv0.6.2",
+    "recipe_node_shape": "BM.GPU.B4.8",
+    "recipe_container_env": [
+        {
+            "key": "tensor_parallel_size",
+            "value": "8"
+        },
+        {
+            "key": "model_name",
+            "value": "NousResearch/Meta-Llama-3.1-70B-Instruct"
+        }
+    ],
+    "recipe_replica_count": 1,
+    "recipe_container_port": "8000",
+    "recipe_nvidia_gpu_count": 8,
+    "recipe_node_pool_size": 1,
+    "recipe_node_boot_volume_size_in_gbs": 200,
+    "recipe_container_command_args": [
+        "--model",
+        "$(model_name)",
+        "--tensor-parallel-size",
+        "$(tensor_parallel_size)"
+    ],
+    "recipe_ephemeral_storage_size": 2000,
+    "recipe_shared_memory_volume_size_limit_in_mb": 80000
+}
diff --git a/docs/sample_recipes/vllm-open-access-hf-model-small b/docs/sample_recipes/vllm-open-access-hf-model-small
new file mode 100644
index 0000000..85b61cd
--- /dev/null
+++ b/docs/sample_recipes/vllm-open-access-hf-model-small
@@ -0,0 +1,30 @@
+{
+    "recipe_id": "llm_inference_nvidia",
+    "recipe_mode": "service",
+    "deployment_name": "vllm-from-hf-small",
+    "recipe_image_uri": "iad.ocir.io/iduyx1qnmway/corrino-devops-repository:vllmv0.6.2",
+    "recipe_node_shape": "VM.GPU.A10.2",
+    "recipe_container_env": [
+        {
+            "key": "tensor_parallel_size",
+            "value": "2"
+        },
+        {
+            "key": "model_name",
+            "value": "NousResearch/Meta-Llama-3-8B-Instruct"
+        }
+    ],
+    "recipe_replica_count": 1,
+    "recipe_container_port": "8000",
+    "recipe_nvidia_gpu_count": 2,
+    "recipe_node_pool_size": 1,
+    "recipe_node_boot_volume_size_in_gbs": 200,
+    "recipe_container_command_args": [
+        "--model",
+        "$(model_name)",
+        "--tensor-parallel-size",
+        "$(tensor_parallel_size)"
+    ],
+    "recipe_ephemeral_storage_size": 100,
+    "recipe_shared_memory_volume_size_limit_in_mb": 200
+}