@@ -7,7 +7,7 @@ schema: 2
 base:
   model:
     path: nvidia/GLM-5.2-NVFP4
-    container: nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc26.dev202609040000
+    container: nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc28.dev202609210001
     precision: fp4
   dynamo:
     install: true
@@ -20,7 +20,11 @@ base:
     gpus_per_node: 4
   engine:
     type: trtllm
-    publish_events_and_metrics: false
+    # Publish engine metrics without KV events. The pinned Dynamo wheel also
+    # selects the routed node address for metadata and transport advertisement
+    # (ai-dynamo/dynamo#15628).
+    # srtctl advertises the worker leaders' /metrics to AIPerf.
+    publish_metrics: true
   roles:
     prefill:
       gpus: 4
@@ -57,7 +61,13 @@ base:
         cache_transceiver_config:
           backend: NIXL
           transceiver_runtime: PYTHON
-          kv_cache_bounce_size_mb: 5120
+          # C++ transfer-agent bounce: every prefill->decode copy is staged in fixed pieces instead of
+          # first-touch writes into the remote pool.
+          agent_bounce_buffer_enable: true
+          agent_bounce_params:
+            max_average_descriptor_size: 4MB
+            min_descriptor_count: '1'
+          kv_cache_bounce_size_mb: 1024
           max_tokens_in_buffer: 1048576
           kv_transfer_timeout_ms: 600000
         cuda_graph_config:
@@ -80,6 +90,7 @@ base:
         moe_expert_parallel_size: 4
         num_postprocess_workers: 8
         pipeline_parallel_size: 1
+        enable_iter_perf_stats: true
         print_iter_log: true
         return_perf_metrics: false
         scheduler_config:
@@ -122,7 +133,13 @@ base:
         cache_transceiver_config:
           backend: NIXL
           transceiver_runtime: PYTHON
-          kv_cache_bounce_size_mb: 5120
+          # C++ transfer-agent bounce: every prefill->decode copy is staged in fixed pieces instead of
+          # first-touch writes into the remote pool.
+          agent_bounce_buffer_enable: true
+          agent_bounce_params:
+            max_average_descriptor_size: 4MB
+            min_descriptor_count: '1'
+          kv_cache_bounce_size_mb: 1024
           max_tokens_in_buffer: 1048576
           kv_transfer_timeout_ms: 600000
         cuda_graph_config:
@@ -142,6 +159,7 @@ base:
           use_low_precision_moe_combine: true
         num_postprocess_workers: 4
         pipeline_parallel_size: 1
+        enable_iter_perf_stats: true
         print_iter_log: true
         return_perf_metrics: false
         sparse_attention_config:
@@ -178,6 +196,7 @@ base:
     command: bash /infmax-workspace/benchmarks/srt_agentic.sh
     env:
       INFMAX_CONTAINER_WORKSPACE: /infmax-workspace
+      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
       RESULT_DIR: /logs/agentic
       PORT: '8000'
       IS_MULTINODE: 'true'
@@ -199,7 +218,7 @@ override_1p1d_tp8_c1_b1_mtp5:
   name: dynamo-disagg-gb300-1p1d-tp8-c1-b1-mtp5
   dynamo:
     source:
-      wheel: 1.4.0.dev20260807
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 1
@@ -225,14 +244,11 @@ override_1p1d_tep8_c20_b5_mtp5:
   name: dynamo-disagg-gb300-1p1d-tep8-c20-b5-mtp5
   dynamo:
     source:
-      git: https://github.com/cquil11/dynamo.git
-      rev: 2cbbdc863c336c2cfffa2a3668e2a6661801b259
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 1
       workers: 1
-      args:
-        enable_iter_perf_stats: true
     decode:
       nodes: 2
       gpus: 8
@@ -245,26 +261,19 @@ override_1p1d_tep8_c20_b5_mtp5:
             - 5
         max_batch_size: 5
         moe_expert_parallel_size: 8
-        enable_iter_perf_stats: true
         sparse_attention_config:
           use_cute_dsl_topk: true
         tensor_parallel_size: 8
-  benchmark:
-    env:
-      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
 
 override_1p4d_tep4_c30_b2_mtp5:
   name: dynamo-disagg-gb300-1p4d-tep4-c30-b2-mtp5
   dynamo:
     source:
-      git: https://github.com/cquil11/dynamo.git
-      rev: 2cbbdc863c336c2cfffa2a3668e2a6661801b259
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 1
       workers: 1
-      args:
-        enable_iter_perf_stats: true
     decode:
       nodes: 4
       workers: 4
@@ -276,26 +285,19 @@ override_1p4d_tep4_c30_b2_mtp5:
             - 2
         max_batch_size: 2
         moe_expert_parallel_size: 4
-        enable_iter_perf_stats: true
         sparse_attention_config:
           use_cute_dsl_topk: true
         tensor_parallel_size: 4
-  benchmark:
-    env:
-      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
 
 override_3p4d_tep4_c60_b5_mtp5:
   name: dynamo-disagg-gb300-3p4d-tep4-c60-b5-mtp5
   dynamo:
     source:
-      git: https://github.com/cquil11/dynamo.git
-      rev: 2cbbdc863c336c2cfffa2a3668e2a6661801b259
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 3
       workers: 3
-      args:
-        enable_iter_perf_stats: true
     decode:
       nodes: 4
       workers: 4
@@ -309,26 +311,20 @@ override_3p4d_tep4_c60_b5_mtp5:
             - 5
         max_batch_size: 5
         moe_expert_parallel_size: 4
-        enable_iter_perf_stats: true
         sparse_attention_config:
           use_cute_dsl_topk: true
         tensor_parallel_size: 4
-  benchmark:
-    env:
-      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
 
 override_4p1d_dep8_c227_b16_mtp3:
   name: dynamo-disagg-gb300-4p1d-dep8-c227-b16-mtp3
   dynamo:
     source:
-      git: https://github.com/cquil11/dynamo.git
-      rev: 2cbbdc863c336c2cfffa2a3668e2a6661801b259
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 4
       workers: 4
       args:
-        enable_iter_perf_stats: true
         speculative_config:
           max_draft_len: 3
     decode:
@@ -351,26 +347,20 @@ override_4p1d_dep8_c227_b16_mtp3:
         moe_config:
           backend: CUTEDSL
         moe_expert_parallel_size: 8
-        enable_iter_perf_stats: true
         speculative_config:
           max_draft_len: 3
         tensor_parallel_size: 8
-  benchmark:
-    env:
-      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
 
 override_5p1d_dep16_c260_b16_mtp3:
   name: dynamo-disagg-gb300-5p1d-dep16-c260-b16-mtp3
   dynamo:
     source:
-      git: https://github.com/cquil11/dynamo.git
-      rev: 2cbbdc863c336c2cfffa2a3668e2a6661801b259
+      wheel: 1.6.0.dev20261010
   roles:
     prefill:
       nodes: 5
       workers: 5
       args:
-        enable_iter_perf_stats: true
         speculative_config:
           max_draft_len: 3
     decode:
@@ -390,12 +380,8 @@ override_5p1d_dep16_c260_b16_mtp3:
           free_gpu_memory_fraction: 0.9
         max_batch_size: 16
         moe_expert_parallel_size: 16
-        enable_iter_perf_stats: true
         sparse_attention_config:
           use_cute_dsl_topk: true
         speculative_config:
           max_draft_len: 3
         tensor_parallel_size: 16
-  benchmark:
-    env:
-      AIPERF_REQUIRED_SERVER_METRIC_PREFIX: trtllm_kv_cache_utilization
@@ -7885,13 +7885,13 @@ glm5.2-fp4-gb300-dynamo-sglang-agentic-disagg:
 # GLM-5.2 NVFP4 TensorRT-LLM AgentX on GB300 with Dynamo disaggregated serving.
 # Six topology variants use NIXL KV transfer and MTP3/MTP5 decoding.
 glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
-  image: nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc26.dev202609040000
+  image: nvcr.io/nvidia/tensorrt-llm/release:1.3.0rc28.dev202609210001
   model: nvidia/GLM-5.2-NVFP4
   model-prefix: glm5.2
   runner: cluster:gb300-nv
   precision: fp4
   framework: dynamo-trt
-  router: { name: dynamo-router, version: "1.4.0.dev20260807" }
+  router: { name: dynamo-router, version: "1.6.0.dev20261010" }
   kv-p2p-transfer: nixl
   multinode: true
   disagg: true
@@ -7902,7 +7902,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [1]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 1
           tp: 4
@@ -7919,7 +7919,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [30]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 1
           tp: 4
@@ -7936,7 +7936,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [20]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 1
           tp: 4
@@ -7953,7 +7953,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [60]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 3
           tp: 4
@@ -7970,7 +7970,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [260]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 5
           tp: 4
@@ -7987,7 +7987,7 @@ glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp:
       - spec-decoding: mtp
         conc-list: [227]
         kv-offloading: dram
-        kv-offload-backend: { name: native, version: "1.3.0rc26.dev202609040000" }
+        kv-offload-backend: { name: native, version: "1.3.0rc28.dev202609210001" }
         prefill:
           num-worker: 4
           tp: 4
 
@@ -9611,3 +9611,13 @@
     - "Concurrency 1 (dcp1 and dcp8) adds quantization-config linear fp8_per_channel: vLLM quantizes the target's linear layers to FP8 W8A8 at load, except kv_b_proj, in_proj_qkvgfab, f_b_proj, fused_qkv_a_proj, self_attn.g_proj and layer 0 mlp.down_proj, which stay BF16. The routed experts stay MXFP4."
     - "No change to the Inferact/Kimi-K3-DSpark draft's precision: its model config takes the speculative config's own quantization (unset) and never receives quantization-config, and get_draft_quant_config() returns None for K3DSparkModel, so its weights stay BF16 and its KV cache stays fp8."
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/3833
+
+- config-keys:
+    - glm5.2-fp4-gb300-dynamo-trt-agentic-disagg-mtp
+  scenario-type:
+    - agentic-coding
+  description:
+    - "Update the GLM-5.2 NVFP4 GB300 TensorRT-LLM AgentX image from 1.3.0rc26.dev202609040000 to 1.3.0rc28.dev202609210001 on all six points."
+    - "Switch NIXL KV transfer to the C++ transfer-agent bounce buffer (max_average_descriptor_size 4MB, min_descriptor_count 1) and lower kv_cache_bounce_size_mb from 5120 to 1024."
+    - "Move Dynamo on all six points to the ai-dynamo 1.6.0.dev20261010 wheel, which includes ai-dynamo/dynamo#15628 (routed node address selection for metadata and transport advertisement)."
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/3707