feat: use generic image and use single node for oss-gpt-120b recipe (#3454)

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

@@ -30,12 +30,12 @@ spec:
           command:
           - /bin/sh
           - -c
-          image: nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.5.1-rc0.pre3
+          image: my-registry/trtllm-runtime:my-tag
       pvc:
         create: false
         mountPoint: /model-store
         name: model-cache
-      replicas: 18
+      replicas: 1
     TrtllmWorker:
       componentType: main
       dynamoNamespace: gpt-oss-agg
@@ -69,7 +69,7 @@ spec:
           command:
           - /bin/sh
           - -c
-          image: nvcr.io/nvidia/ai-dynamo/tensorrtllm-runtime:0.5.1-rc0.pre3
+          image: my-registry/trtllm-runtime:my-tag
           env:
           - name: TRTLLM_ENABLE_PDL
             value: "1"
@@ -80,7 +80,7 @@ spec:
           - name: ENGINE_ARGS
             value: "/opt/dynamo/configs/config.yaml"
           - name: MODEL_PATH
-            value: "/model-store/models--openai--gpt-oss-120b/snapshots/b5c939de8f754692c1647ca79fbf85e8c1e70f8a"
+            value: "/model-store/hub/models--openai--gpt-oss-120b/snapshots/b5c939de8f754692c1647ca79fbf85e8c1e70f8a"
           volumeMounts:
           - mountPath: /opt/dynamo/configs
             name: llm-config
@@ -94,7 +94,7 @@ spec:
         create: false
         mountPoint: /model-store
         name: model-cache
-      replicas: 18
+      replicas: 1
       resources:
         limits:
           gpu: "4"

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

@@ -114,7 +114,7 @@ spec:
         - name: CONCURRENCY_PER_GPU
           value: "900"
         - name: DEPLOYMENT_GPU_COUNT
-          value: "72"
+          value: "4"
         - name: ISL
           value: "128"
         - name: OSL