fix: misnamed tensorrtllm-runtime image and incorrect tag (#4289)

Signed-off-by: Harrison King Saturley-Hall <hsaturleyhal@nvidia.com>
Signed-off-by: Harrison Saturley-Hall <harrison.saturley.hall@gmail.com>
Co-authored-by: hhzhang16 <54051230+hhzhang16@users.noreply.github.com>

benchmarks/profiler/deploy/profile_sla_aic_dgdr.yaml

@@ -12,7 +12,7 @@ spec:
 
   # ProfilingConfig maps directly to the profile_sla.py config format
   profilingConfig:
-    profilerImage: "nvcr.io/nvidian/dynamo-dev/vllm-runtime:dep-554.0"
+    profilerImage: "nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag"
     config:
       # Sweep/profiling configuration
       sweep:
@@ -31,7 +31,7 @@ spec:
 
   # Deployment overrides for the auto-created DGD
   deploymentOverrides:
-    workersImage: "nvcr.io/nvidian/dynamo-dev/trtllm-runtime:dep-554.0"
+    workersImage: "nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:my-tag"
 
   # Automatically create DynamoGraphDeployment after profiling
   autoApply: true

benchmarks/profiler/deploy/profile_sla_dgdr.yaml

@@ -12,7 +12,7 @@ spec:
 
   # ProfilingConfig maps directly to the profile_sla.py config format
   profilingConfig:
-    profilerImage: "nvcr.io/nvidian/dynamo-dev/vllm-runtime:dep-554.0"
+    profilerImage: "nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag"
     config:
       # Sweep/profiling configuration
       sweep:
@@ -28,7 +28,7 @@ spec:
 
   # Deployment overrides for the auto-created DGD
   deploymentOverrides:
-    workersImage: "nvcr.io/nvidian/dynamo-dev/vllm-runtime:dep-554.0"
+    workersImage: "nvcr.io/nvidia/ai-dynamo/vllm-runtime:my-tag"
 
   # Automatically create DynamoGraphDeployment after profiling
   autoApply: true

deploy/cloud/operator/config/samples/nvidia.com_v1alpha1_dynamographdeploymentrequest.yaml

@@ -25,7 +25,7 @@ spec:
   backend: trtllm
 
   # ProfilerImage is the container image to use for profiling jobs (required)
-  profilerImage: "nvcr.io/nvidia/ai-dynamo/trtllm-runtime:0.6.1"
+  profilerImage: "nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.6.1"
 
   # ProfilingConfig maps directly to the profile_sla.py config format
   # See benchmarks/profiler/utils/profiler_argparse.py for complete schema

docs/benchmarks/sla_driven_profiling.md

@@ -392,7 +392,7 @@ spec:
   backend: trtllm
 
   profilingConfig:
-    profilerImage: "nvcr.io/nvidia/ai-dynamo/trtllm-runtime:0.6.1"
+    profilerImage: "nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.6.1"
     config:
       sla:
         isl: 4000
@@ -409,7 +409,7 @@ spec:
         backend_version: "0.20.0"
 
   deploymentOverrides:
-    workersImage: "nvcr.io/nvidia/ai-dynamo/trtllm-runtime:0.6.1"
+    workersImage: "nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.6.1"
 
   autoApply: true
 ```
@@ -493,7 +493,7 @@ AssertionError: num_heads <N> should be divisible by tp_size <M> and the divisio
 
 ```yaml
 profilingConfig:
-  profilerImage: "nvcr.io/nvidia/ai-dynamo/trtllm-runtime:0.6.1"
+  profilerImage: "nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.6.1"
   config:
     hardware:
       max_num_gpus_per_engine: 4  # For Qwen3-0.6B (16 heads / 4 = max TP of 4)

examples/backends/trtllm/deploy/README.md

@@ -89,7 +89,7 @@ resources:
 ```yaml
 extraPodSpec:
   mainContainer:
-    image: my-registry/trtllm-runtime:my-tag
+    image: my-registry/tensorrtllm-runtime:my-tag
     workingDir: /workspace/examples/backends/trtllm
     args:
       - "python3"
@@ -109,7 +109,7 @@ Before using these templates, ensure you have:
 
 ### Container Images
 
-The deployment files currently require access to `my-registry/trtllm-runtime`. If you don't have access, build and push your own image:
+The deployment files currently require access to `my-registry/tensorrtllm-runtime`. If you don't have access, build and push your own image:
 
 ```bash
 ./container/build.sh --framework tensorrtllm
@@ -141,7 +141,7 @@ Edit the template to match your environment:
 
 ```yaml
 # Update image registry and tag
-image: my-registry/trtllm-runtime:my-tag
+image: my-registry/tensorrtllm-runtime:my-tag
 
 # Configure your model and deployment settings
 args:

examples/backends/trtllm/deploy/agg.yaml

@@ -13,7 +13,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
     TRTLLMWorker:
       envFromSecret: hf-token-secret
       dynamoNamespace: trtllm-agg
@@ -24,7 +24,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3

examples/backends/trtllm/deploy/agg_router.yaml

@@ -13,7 +13,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
       envs:
         - name: DYN_ROUTER_MODE
           value: kv
@@ -27,7 +27,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3

examples/backends/trtllm/deploy/disagg-multinode.yaml

@@ -95,7 +95,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/examples/backends/trtllm
           command:
           - python3
@@ -127,7 +127,7 @@ spec:
             - name: nvidia-config
               mountPath: /workspace/
               readOnly: true
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3
@@ -165,7 +165,7 @@ spec:
             - name: nvidia-config
               mountPath: /workspace/
               readOnly: true
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3

examples/backends/trtllm/deploy/disagg.yaml

@@ -13,7 +13,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
     TRTLLMPrefillWorker:
       dynamoNamespace: trtllm-disagg
       envFromSecret: hf-token-secret
@@ -25,7 +25,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3
@@ -51,7 +51,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3

examples/backends/trtllm/deploy/disagg_planner.yaml

@@ -13,7 +13,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/examples/backends/trtllm
           command:
             - python3
@@ -38,7 +38,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/components/src/dynamo/planner
           ports:
             - name: metrics
@@ -89,7 +89,7 @@ spec:
       extraPodSpec:
         terminationGracePeriodSeconds: 600
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
             - python3
@@ -116,7 +116,7 @@ spec:
       extraPodSpec:
         terminationGracePeriodSeconds: 600
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
             - python3

examples/backends/trtllm/deploy/disagg_router.yaml

@@ -13,7 +13,7 @@ spec:
       replicas: 1
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
       envs:
         - name: DYN_ROUTER_MODE
           value: kv
@@ -27,7 +27,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3
@@ -53,7 +53,7 @@ spec:
           gpu: "1"
       extraPodSpec:
         mainContainer:
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           workingDir: /workspace/
           command:
           - python3

recipes/README.md

@@ -74,7 +74,7 @@ Ensure your Kubernetes cluster has:
 
 Ensure access to NVIDIA container registry for runtime images:
 - `nvcr.io/nvidia/ai-dynamo/vllm-runtime:x.y.z`
-- `nvcr.io/nvidia/ai-dynamo/trtllm-runtime:x.y.z`
+- `nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:x.y.z`
 - `nvcr.io/nvidia/ai-dynamo/sglang-runtime:x.y.z`
 
 ### 5. HuggingFace Access and Kubernetes Secret Creation

recipes/gpt-oss-120b/trtllm/agg/deploy.yaml

@@ -46,7 +46,7 @@ spec:
           command:
           - /bin/sh
           - -c
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
       replicas: 1
     TrtllmWorker:
       componentType: main
@@ -81,7 +81,7 @@ spec:
           command:
           - /bin/sh
           - -c
-          image: my-registry/trtllm-runtime:my-tag
+          image: my-registry/tensorrtllm-runtime:my-tag
           env:
           - name: TRTLLM_ENABLE_PDL
             value: "1"