feat: support AIC DGD gen call (WILL BREAK DGDR) (#6216)

Signed-off-by: hongkuanz <hongkuanz@nvidia.com>

docs/pages/components/planner/README.md

@@ -42,7 +42,7 @@ The Planner monitors system performance and automatically scales prefill/decode
 The fastest path to a planner-enabled deployment is through a DynamoGraphDeploymentRequest:
 
 ```bash
-kubectl apply -f benchmarks/profiler/deploy/profile_sla_aic_dgdr.yaml -n $NAMESPACE
+kubectl apply -f components/src/dynamo/profiler/deploy/profile_sla_aic_dgdr.yaml -n $NAMESPACE
 ```
 
 This automatically profiles your model and deploys with the SLA planner. See [SLA Planner Guide](planner-guide.md) for the full workflow.

docs/pages/components/planner/planner-examples.md

@@ -45,7 +45,7 @@ spec:
 Deploy:
 ```bash
 export NAMESPACE=your-namespace
-kubectl apply -f benchmarks/profiler/deploy/profile_sla_aic_dgdr.yaml -n $NAMESPACE
+kubectl apply -f components/src/dynamo/profiler/deploy/profile_sla_aic_dgdr.yaml -n $NAMESPACE
 ```
 
 ### Online Profiling (Real Measurements)
@@ -82,10 +82,10 @@ spec:
 
 Deploy:
 ```bash
-kubectl apply -f benchmarks/profiler/deploy/profile_sla_dgdr.yaml -n $NAMESPACE
+kubectl apply -f components/src/dynamo/profiler/deploy/profile_sla_dgdr.yaml -n $NAMESPACE
 ```
 
-Available sample DGDRs in `benchmarks/profiler/deploy/`:
+Available sample DGDRs in `components/src/dynamo/profiler/deploy/`:
 - **`profile_sla_dgdr.yaml`**: Standard online profiling for dense models
 - **`profile_sla_aic_dgdr.yaml`**: Fast offline profiling using AI Configurator
 - **`profile_sla_moe_dgdr.yaml`**: Online profiling for MoE models (SGLang)
@@ -126,7 +126,7 @@ spec:
 
 Deploy:
 ```bash
-kubectl apply -f benchmarks/profiler/deploy/profile_sla_moe_dgdr.yaml -n $NAMESPACE
+kubectl apply -f components/src/dynamo/profiler/deploy/profile_sla_moe_dgdr.yaml -n $NAMESPACE
 ```
 
 ### Using Existing DGD Configs (Custom Setups)

docs/pages/components/profiler/README.md

@@ -77,7 +77,7 @@ profilingConfig:
 For advanced scenarios, run the profiler directly:
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend vllm \
   --config path/to/disagg.yaml \
   --model meta-llama/Llama-3-8B \

docs/pages/components/profiler/profiler-examples.md

@@ -162,7 +162,7 @@ spec:
 Launch an interactive configuration selection interface:
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend trtllm \
   --config path/to/disagg.yaml \
   --pick-with-webui \
@@ -224,7 +224,7 @@ Once you select a configuration, the full DGD CRD is saved as `config_with_plann
 ### Basic Profiling
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend vllm \
   --config path/to/disagg.yaml \
   --model meta-llama/Llama-3-8B \
@@ -235,7 +235,7 @@ python -m benchmarks.profiler.profile_sla \
 ### With GPU Constraints
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend sglang \
   --config examples/backends/sglang/deploy/disagg.yaml \
   --model deepseek-ai/DeepSeek-R1-Distill-Llama-8B \
@@ -248,7 +248,7 @@ python -m benchmarks.profiler.profile_sla \
 ### AI Configurator (Offline)
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend trtllm \
   --config path/to/disagg.yaml \
   --use-ai-configurator \

docs/pages/components/profiler/profiler-guide.md

@@ -50,7 +50,7 @@ The profiler sweeps over the following parallelization mappings for prefill and
 
 ### Kubernetes Deployment (DGDR)
 
-The recommended deployment method is through DGDRs. Sample configurations are provided in `benchmarks/profiler/deploy/`:
+The recommended deployment method is through DGDRs. Sample configurations are provided in `components/src/dynamo/profiler/deploy/`:
 
 | Sample | Description |
 |--------|-------------|
@@ -148,7 +148,7 @@ curl http://localhost:8000/v1/models
 For advanced use cases or local development:
 
 ```bash
-python -m benchmarks.profiler.profile_sla \
+python -m dynamo.profiler.profile_sla \
   --backend vllm \
   --config path/to/disagg.yaml \
   --model meta-llama/Llama-3-8B \
@@ -644,4 +644,4 @@ kubectl create secret docker-registry nvcr-imagepullsecret \
 - [SLA Planner Guide](../planner/planner-guide.md) - End-to-end deployment workflow
 - [SLA Planner Architecture](../planner/planner-guide.md) - How the Planner uses profiling data
 - [DGDR API Reference](../../kubernetes/api-reference.md) - DGDR specification
-- [Profiler Arguments Reference](https://github.com/ai-dynamo/dynamo/blob/main/benchmarks/profiler/utils/profiler_argparse.py) - Full CLI reference
+- [Profiler Arguments Reference](https://github.com/ai-dynamo/dynamo/blob/main/components/src/dynamo/profiler/utils/profiler_argparse.py) - Full CLI reference

docs/pages/kubernetes/api-reference.md

@@ -858,7 +858,7 @@ _Appears in:_
 
 ProfilingConfigSpec defines configuration for the profiling process.
 This structure maps directly to the profile_sla.py config format.
-See benchmarks/profiler/utils/profiler_argparse.py for the complete schema.
+See dynamo/profiler/utils/profiler_argparse.py for the complete schema.
 
 
 

tests/profiler/test_profile_sla_aiconfigurator.py

@@ -17,8 +17,8 @@ import pytest
 project_root = Path(__file__).parent.parent.parent
 sys.path.insert(0, str(project_root))
 
-from benchmarks.profiler.profile_sla import run_profile  # noqa: E402
-from benchmarks.profiler.utils.model_info import ModelInfo  # noqa: E402
+from dynamo.profiler.profile_sla import run_profile  # noqa: E402
+from dynamo.profiler.utils.model_info import ModelInfo  # noqa: E402
 
 pytestmark = [
     pytest.mark.aiconfigurator,

tests/profiler/test_profile_sla_dryrun.py

@@ -18,9 +18,9 @@ import pytest
 project_root = Path(__file__).parent.parent.parent
 sys.path.insert(0, str(project_root))
 
-from benchmarks.profiler.profile_sla import run_profile  # noqa: E402
-from benchmarks.profiler.utils.model_info import ModelInfo  # noqa: E402
-from benchmarks.profiler.utils.search_space_autogen import (  # noqa: E402
+from dynamo.profiler.profile_sla import run_profile  # noqa: E402
+from dynamo.profiler.utils.model_info import ModelInfo  # noqa: E402
+from dynamo.profiler.utils.search_space_autogen import (  # noqa: E402
     auto_generate_search_space,
 )
 
@@ -340,8 +340,8 @@ class TestProfileSLADryRun:
     @pytest.mark.integration
     @pytest.mark.gpu_0
     @pytest.mark.vllm
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_gpu_summary")
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_model_info")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_gpu_summary")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_model_info")
     async def test_profile_with_autogen_search_space_h100(
         self,
         mock_get_model_info,
@@ -411,8 +411,8 @@ class TestProfileSLADryRun:
     @pytest.mark.gpu_0
     @pytest.mark.integration
     @pytest.mark.sglang
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_gpu_summary")
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_model_info")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_gpu_summary")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_model_info")
     async def test_sglang_profile_with_autogen_search_space_h100(
         self,
         mock_get_model_info,
@@ -482,8 +482,8 @@ class TestProfileSLADryRun:
     @pytest.mark.gpu_0
     @pytest.mark.integration
     @pytest.mark.trtllm
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_gpu_summary")
-    @patch("benchmarks.profiler.utils.search_space_autogen.get_model_info")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_gpu_summary")
+    @patch("dynamo.profiler.utils.search_space_autogen.get_model_info")
     async def test_trtllm_profile_with_autogen_search_space_h100(
         self,
         mock_get_model_info,