diff --git a/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx-agg.yaml b/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx-agg.yaml
new file mode 100644
index 000000000..87979921f
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx-agg.yaml
@@ -0,0 +1,118 @@
+# Agentic-coding SGLang aggregated recipe for GLM-5.2-NVFP4 on GB300
+# (single aggregated worker, TP8, EAGLE MTP + hierarchical-cache KV).
+#
+# Flat single-variant schema the agentic CI flow expects: applied via
+# CONFIG_FILE=<this file> with no zip_override selector, so it must NOT be
+# base:-wrapped. Concurrency is not a recipe field; the GHA matrix fans out
+# one job per concurrency from the master-config conc-list into agentic_srt.sh.
+name: gb300-fp4-glm5.2-agentx-agg
+model:
+  path: glm-5.2-fp4
+  container: dynamo-sglang
+  precision: fp4
+resources:
+  gpu_type: gb300
+  gpus_per_node: 4
+  agg_nodes: 2
+  agg_workers: 1
+  gpus_per_agg: 8
+frontend:
+  type: dynamo
+  env:
+    PIP_BREAK_SYSTEM_PACKAGES: '1'
+  args:
+    router-mode: kv
+    active-decode-blocks-threshold: None
+    active-prefill-tokens-threshold: None
+    active-prefill-tokens-threshold-frac: None
+    router-session-affinity-ttl-secs: 3600
+dynamo:
+  hash: 71eb001e17fa73c742f0afe1a6ed96836cb135fd
+  install: true
+backend:
+  type: sglang
+  sglang_config:
+    aggregated:
+      served-model-name: nvidia/GLM-5.2-NVFP4
+      trust-remote-code: true
+      quantization: modelopt_fp4
+      kv-cache-dtype: fp8_e4m3
+      tensor-parallel-size: 8
+      data-parallel-size: 1
+      expert-parallel-size: 1
+      enable-dp-attention: false
+      enable-dp-lm-head: false
+      max-running-requests: 10
+      cuda-graph-max-bs: 10
+      chunked-prefill-size: 8192
+      max-prefill-tokens: 8192
+      context-length: 1048576
+      speculative-algorithm: EAGLE
+      speculative-num-steps: 4
+      speculative-eagle-topk: 1
+      speculative-num-draft-tokens: 5
+      nsa-prefill-backend: trtllm
+      nsa-decode-backend: trtllm
+      moe-runner-backend: flashinfer_trtllm
+      fp4-gemm-backend: flashinfer_trtllm
+      disable-shared-experts-fusion: true
+      enable-flashinfer-allreduce-fusion: true
+      weight-loader-prefetch-checkpoints: true
+      model-loader-extra-config: '{"enable_multithread_load": true}'
+      mem-fraction-static: 0.8
+      enable-hierarchical-cache: true
+      hicache-write-policy: write_back
+      hicache-size: 135
+      hicache-io-backend: direct
+      enable-metrics: true
+      enable-cache-report: true
+  aggregated_environment:
+    TORCH_DISTRIBUTED_DEFAULT_TIMEOUT: '1800'
+    PYTHONUNBUFFERED: '1'
+    DYN_SKIP_SGLANG_LOG_FORMATTING: '1'
+    MC_TE_METRIC: 'true'
+    NCCL_CUMEM_ENABLE: '1'
+    SGLANG_USE_MESSAGE_QUEUE_BROADCASTER: '0'
+    SGLANG_DISABLE_TP_MEMORY_INBALANCE_CHECK: '1'
+    SGLANG_ENABLE_THINKING: '1'
+    SGLANG_REASONING_EFFORT: max
+    SGLANG_ENABLE_UNIFIED_RADIX_TREE: '1'
+    SGLANG_HICACHE_DEBUG_LOG: '1'
+    SGLANG_HICACHE_DEBUG_SAMPLE_RATE: '16384'
+    SGLANG_MOE_NVFP4_DISPATCH: '1'
+    SGLANG_NVFP4_CKPT_FP8_NEXTN_MOE: '1'
+    SGLANG_CLIP_MAX_NEW_TOKENS_ESTIMATION: '8'
+    SGLANG_SIMULATE_ACC_LEN: '3.33'
+    SGLANG_SIMULATE_ACC_METHOD: match-expected
+    SGLANG_SIMULATE_ACC_TOKEN_MODE: real-draft-token
+    PIP_BREAK_SYSTEM_PACKAGES: '1'
+    SGLANG_DG_CACHE_DIR: /deepgemm_cache
+    FLASHINFER_WORKSPACE_BASE: /flashinfer_cache
+    MC_FORCE_MNNVL: '1'
+    NCCL_MNNVL_ENABLE: '1'
+    SGLANG_MOONCAKE_CUSTOM_MEM_POOL: 'True'
+    UCX_TLS: ib,cuda_copy,cuda_ipc,sm,self
+    NVSHMEM_REMOTE_TRANSPORT: none
+health_check:
+  max_attempts: 1440
+  interval_seconds: 10
+benchmark:
+  type: custom
+  command: bash /infmax-workspace/benchmarks/multi_node/agentic_srt.sh
+  env:
+    INFMAX_CONTAINER_WORKSPACE: /infmax-workspace
+    RESULT_DIR: /logs/agentic
+    PORT: "8000"
+    IS_MULTINODE: "true"
+    AIPERF_HTTP_X_DYNAMO_SESSION_ID_FROM_CORRELATION_ID: "true"
+    AIPERF_DATASET_MMAP_CACHE_DIR: /aiperf_mmap_cache
+    HF_HUB_CACHE: /hf_hub_cache
+infra:
+  etcd_nats_dedicated_node: false
+  nats_max_payload_mb: 64
+sbatch_directives:
+  mem: '0'
+  cpus-per-task: '144'
+srun_options:
+  mem: '0'
+  container-remap-root: ''
diff --git a/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml b/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml
new file mode 100644
index 000000000..b04ea1752
--- /dev/null
+++ b/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml
@@ -0,0 +1,219 @@
+base:
+  name: gb300-fp4-glm5.2-agentx
+  model:
+    path: glm-5.2-fp4
+    container: dynamo-sglang
+    precision: fp4
+  resources:
+    gpu_type: gb300
+    gpus_per_node: 4
+  frontend:
+    type: dynamo
+    nginx_session_affinity: true
+    nginx_session_affinity_header: X-Dynamo-Session-ID
+    enable_multiple_frontends: true
+    env:
+      PIP_BREAK_SYSTEM_PACKAGES: '1'
+    args:
+      router-mode: kv
+      active-decode-blocks-threshold: None
+      active-prefill-tokens-threshold: None
+      active-prefill-tokens-threshold-frac: None
+      router-session-affinity-ttl-secs: 3600
+  dynamo:
+    hash: 71eb001e17fa73c742f0afe1a6ed96836cb135fd
+    install: true
+  backend:
+    type: sglang
+    prefill_environment:
+      TORCH_DISTRIBUTED_DEFAULT_TIMEOUT: '1800'
+      PYTHONUNBUFFERED: '1'
+      DYN_SKIP_SGLANG_LOG_FORMATTING: '1'
+      SGLANG_DISAGGREGATION_HEARTBEAT_MAX_FAILURE: '100000'
+      SGLANG_DISAGGREGATION_BOOTSTRAP_TIMEOUT: '100000'
+      SGLANG_DISAGGREGATION_WAITING_TIMEOUT: '100000'
+      MC_TE_METRIC: 'true'
+      MC_FORCE_MNNVL: '1'
+      NCCL_MNNVL_ENABLE: '1'
+      NCCL_CUMEM_ENABLE: '1'
+      SGLANG_MOONCAKE_CUSTOM_MEM_POOL: 'True'
+      SGLANG_USE_MESSAGE_QUEUE_BROADCASTER: '0'
+      SGLANG_DISABLE_TP_MEMORY_INBALANCE_CHECK: '1'
+      SGLANG_ENABLE_THINKING: '1'
+      SGLANG_ENABLE_UNIFIED_RADIX_TREE: '1'
+      SGLANG_HICACHE_DEBUG_LOG: '1'
+      SGLANG_HICACHE_DEBUG_SAMPLE_RATE: '16384'
+      SGLANG_REASONING_EFFORT: max
+      PIP_BREAK_SYSTEM_PACKAGES: '1'
+      UCX_TLS: ib,cuda_copy,cuda_ipc,sm,self
+      NVSHMEM_REMOTE_TRANSPORT: none
+      SGLANG_DG_CACHE_DIR: /deepgemm_cache
+      FLASHINFER_WORKSPACE_BASE: /flashinfer_cache
+      SGLANG_SIMULATE_ACC_LEN: '2.5'
+    decode_environment:
+      TORCH_DISTRIBUTED_DEFAULT_TIMEOUT: '1800'
+      PYTHONUNBUFFERED: '1'
+      DYN_SKIP_SGLANG_LOG_FORMATTING: '1'
+      SGLANG_DISAGGREGATION_HEARTBEAT_MAX_FAILURE: '100000'
+      SGLANG_DISAGGREGATION_BOOTSTRAP_TIMEOUT: '100000'
+      SGLANG_DISAGGREGATION_WAITING_TIMEOUT: '100000'
+      MC_TE_METRIC: 'true'
+      MC_FORCE_MNNVL: '1'
+      NCCL_MNNVL_ENABLE: '1'
+      NCCL_CUMEM_ENABLE: '1'
+      SGLANG_MOONCAKE_CUSTOM_MEM_POOL: 'True'
+      SGLANG_USE_MESSAGE_QUEUE_BROADCASTER: '0'
+      SGLANG_DISABLE_TP_MEMORY_INBALANCE_CHECK: '1'
+      SGLANG_DEEPEP_NUM_MAX_DISPATCH_TOKENS_PER_RANK: '512'
+      SGLANG_MOE_NVFP4_DISPATCH: '1'
+      SGLANG_NVFP4_CKPT_FP8_NEXTN_MOE: '1'
+      SGLANG_ENABLE_THINKING: '1'
+      SGLANG_CLIP_MAX_NEW_TOKENS_ESTIMATION: '8'
+      SGLANG_REASONING_EFFORT: max
+      PIP_BREAK_SYSTEM_PACKAGES: '1'
+      UCX_TLS: ib,cuda_copy,cuda_ipc,sm,self
+      NVSHMEM_REMOTE_TRANSPORT: none
+      SGLANG_DG_CACHE_DIR: /deepgemm_cache
+      FLASHINFER_WORKSPACE_BASE: /flashinfer_cache
+      SGLANG_SIMULATE_ACC_LEN: '2.5'
+      SGLANG_SIMULATE_ACC_METHOD: match-expected
+      SGLANG_SIMULATE_ACC_TOKEN_MODE: real-draft-token
+    sglang_config:
+      prefill:
+        served-model-name: nvidia/GLM-5.2-NVFP4
+        trust-remote-code: true
+        quantization: modelopt_fp4
+        kv-cache-dtype: fp8_e4m3
+        disaggregation-mode: prefill
+        disaggregation-transfer-backend: nixl
+        tensor-parallel-size: 4
+        data-parallel-size: 4
+        expert-parallel-size: 4
+        enable-dp-attention: true
+        enable-dp-lm-head: true
+        load-balance-method: total_tokens
+        chunked-prefill-size: 65536
+        max-prefill-tokens: 16384
+        max-running-requests: 16
+        cuda-graph-max-bs: 16
+        disable-cuda-graph: true
+        nsa-prefill-backend: trtllm
+        nsa-decode-backend: trtllm
+        moe-runner-backend: flashinfer_cutlass
+        fp4-gemm-backend: flashinfer_cutlass
+        enable-flashinfer-allreduce-fusion: true
+        weight-loader-prefetch-checkpoints: true
+        model-loader-extra-config: '{"enable_multithread_load": true}'
+        mem-fraction-static: 0.8
+        context-length: 1048576
+        enable-hierarchical-cache: true
+        hicache-write-policy: write_back
+        hicache-size: 135
+        hicache-io-backend: direct
+        speculative-algorithm: EAGLE
+        speculative-num-steps: 1
+        speculative-eagle-topk: 1
+        speculative-num-draft-tokens: 2
+        enable-metrics: true
+        enable-cache-report: true
+      decode:
+        served-model-name: nvidia/GLM-5.2-NVFP4
+        trust-remote-code: true
+        quantization: modelopt_fp4
+        kv-cache-dtype: fp8_e4m3
+        disaggregation-mode: decode
+        disaggregation-transfer-backend: nixl
+        disable-radix-cache: true
+        speculative-algorithm: EAGLE
+        speculative-num-steps: 2
+        speculative-eagle-topk: 1
+        speculative-num-draft-tokens: 3
+        tensor-parallel-size: 4
+        data-parallel-size: 1
+        expert-parallel-size: 1
+        enable-dp-attention: false
+        enable-dp-lm-head: false
+        max-running-requests: 16
+        cuda-graph-max-bs: 16
+        chunked-prefill-size: 64
+        context-length: 1048576
+        nsa-prefill-backend: trtllm
+        nsa-decode-backend: trtllm
+        moe-runner-backend: flashinfer_trtllm
+        fp4-gemm-backend: flashinfer_cutlass
+        skip-tokenizer-init: true
+        stream-interval: 30
+        enable-flashinfer-allreduce-fusion: true
+        weight-loader-prefetch-checkpoints: true
+        model-loader-extra-config: '{"enable_multithread_load": true}'
+        mem-fraction-static: 0.9
+        disaggregation-decode-extra-slots: 0
+        enable-metrics: true
+        enable-cache-report: true
+  health_check:
+    max_attempts: 1440
+    interval_seconds: 10
+  benchmark:
+    type: custom
+    command: bash /infmax-workspace/benchmarks/multi_node/agentic_srt.sh
+    env:
+      INFMAX_CONTAINER_WORKSPACE: /infmax-workspace
+      RESULT_DIR: /logs/agentic
+      PORT: "8000"
+      IS_MULTINODE: "true"
+      AIPERF_HTTP_X_DYNAMO_SESSION_ID_FROM_CORRELATION_ID: "true"
+      AIPERF_DATASET_MMAP_CACHE_DIR: /aiperf_mmap_cache
+      HF_HUB_CACHE: /hf_hub_cache
+  infra:
+    etcd_nats_dedicated_node: false
+    nats_max_payload_mb: 64
+  sbatch_directives:
+    mem: '0'
+    cpus-per-task: '144'
+  srun_options:
+    mem: '0'
+    container-remap-root: ''
+# ################# agentx #################
+zip_override_mtp_agentx_hightpt:
+  name: [agentx-2p1d_dep16-c128, agentx-2p1d_dep16-c192]
+  backend:
+    sglang_config:
+      decode:
+        cuda-graph-max-bs: 144
+        data-parallel-size: 16
+        deepep-config: /configs/deepep_config.json
+        deepep-mode: low_latency
+        enable-dp-attention: true
+        enable-dp-lm-head: true
+        ep-dispatch-algorithm: static
+        ep-num-redundant-experts: 0
+        expert-parallel-size: 16
+        max-running-requests: 144
+        mem-fraction-static: 0.85
+        moe-a2a-backend: deepep
+        moe-dense-tp-size: 1
+        moe-runner-backend: flashinfer_cutedsl
+        speculative-moe-a2a-backend: deepep
+        speculative-moe-runner-backend: deep_gemm
+        tensor-parallel-size: 16
+      prefill:
+        data-parallel-size: 8
+        expert-parallel-size: 8
+        max-prefill-tokens: 8192
+        tensor-parallel-size: 8
+  resources:
+    decode_nodes: 4
+    decode_workers: 1
+    gpus_per_decode: 16
+    gpus_per_prefill: 8
+    prefill_nodes: 4
+    prefill_workers: 2
+zip_override_mtp_agentx_lowlat:
+  name: [agentx-1p2d_tp4-c48, agentx-1p4d_tp4-c48, agentx-1p6d_tp4-c45]
+  resources:
+    decode_nodes: [2, 4, 6]
+    decode_workers: [2, 4, 6]
+    gpus_per_decode: 4
+    gpus_per_prefill: 4
+    prefill_nodes: 1
+    prefill_workers: 1
diff --git a/configs/nvidia-master.yaml b/configs/nvidia-master.yaml
index a60792bd6..62ba9cff3 100644
--- a/configs/nvidia-master.yaml
+++ b/configs/nvidia-master.yaml
@@ -7955,8 +7955,7 @@ glm5.2-fp8-h200-dynamo-sglang-agentic-mtp-2p2d:
 # eagle-topk 1, 4 draft tokens = 3 speculative tokens), with acceptance pinned
 # to the golden AL 2.99 (golden_al_distribution/glm5.2_mtp.yaml, thinking_on,
 # K=3) through SGLANG_SIMULATE_ACC_*. Image is bumped to v0.5.16-cu130: it is
-# the first release that reads SGLANG_SIMULATE_ACC_TOKEN_MODE, and it is the
-# tag the GLM-5.2 GB300 dynamo-sglang agentic recipes already run.
+# the first release that reads SGLANG_SIMULATE_ACC_TOKEN_MODE.
 #
 # Cookbook low-latency arm (TP8, fp8 KV, cutedsl bf16 GEMM) only, on the AgentX
 # MTP concurrency grid [1, 4, 8, 12, 16]: steps of at least 2 because
@@ -7989,7 +7988,7 @@ glm5.2-fp4-b300-sglang-agentic-mtp:
 # acceptance pinned to the golden AL 2.99 (golden_al_distribution/glm5.2_mtp.yaml,
 # thinking_on, K=3) through SGLANG_SIMULATE_ACC_*. Image lmsysorg/sglang:v0.5.16-cu130
 # is the first release that reads SGLANG_SIMULATE_ACC_TOKEN_MODE and is what the
-# B300 sibling and the GLM-5.2 GB300 dynamo-sglang agentic recipes run.
+# B300 sibling runs.
 #
 # Same single arm and concurrency grid as glm5.2-fp4-b300-sglang-agentic-mtp so
 # the two SKUs are directly comparable: cookbook low-latency TP8 with HiCache
@@ -8010,6 +8009,140 @@ glm5.2-fp4-b200-sglang-agentic-mtp:
       search-space:
       - { tp: 8, spec-decoding: mtp, kv-offloading: dram, kv-offload-backend: { name: hicache }, conc-list: [1, 4, 8, 12, 16] }
 
+glm5.2-fp4-gb300-dynamo-sglang-agentic-agg:
+  image: lmsysorg/sglang:nightly-dev-cu13-20260805-211ee642
+  model: nvidia/GLM-5.2-NVFP4
+  model-prefix: glm5.2
+  runner: cluster:gb300-nv
+  precision: fp4
+  framework: dynamo-sglang
+  router: { name: dynamo-router, version: "71eb001e17fa73c742f0afe1a6ed96836cb135fd" }
+  multinode: true
+  disagg: false
+  scenarios:
+    agentic-coding:
+    - dram-utilization: 0.80
+      search-space:
+      # aggregated 1-worker TP8 (prefill+decode on the same GPUs)
+      - spec-decoding: mtp
+        conc-list: [2, 4]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 1
+          tp: 8
+          ep: 1
+          dp-attn: false
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx-agg.yaml"
+        # Aggregated worker serves prefill and decode on the same GPUs.
+        # Keep decode at zero to avoid double-counting the TP8 allocation.
+        decode:
+          num-worker: 0
+          tp: 8
+          ep: 1
+          dp-attn: false
+
+glm5.2-fp4-gb300-dynamo-sglang-agentic-disagg:
+  image: lmsysorg/sglang:nightly-dev-cu13-20260805-211ee642
+  model: nvidia/GLM-5.2-NVFP4
+  model-prefix: glm5.2
+  runner: cluster:gb300-nv
+  precision: fp4
+  framework: dynamo-sglang
+  router: { name: dynamo-router, version: "71eb001e17fa73c742f0afe1a6ed96836cb135fd" }
+  kv-p2p-transfer: nixl
+  multinode: true
+  disagg: true
+  scenarios:
+    agentic-coding:
+    - dram-utilization: 0.80
+      search-space:
+      # low-latency 1P2D TP4-prefill / TP4-decode (2 decode workers)
+      - spec-decoding: mtp
+        conc-list: [48]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 4
+          dp-attn: true
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml:zip_override_mtp_agentx_lowlat[0]"
+        decode:
+          num-worker: 2
+          tp: 4
+          ep: 1
+          dp-attn: false
+      # low-latency 1P4D TP4-prefill / TP4-decode (4 decode workers)
+      - spec-decoding: mtp
+        conc-list: [48]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 4
+          dp-attn: true
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml:zip_override_mtp_agentx_lowlat[1]"
+        decode:
+          num-worker: 4
+          tp: 4
+          ep: 1
+          dp-attn: false
+      # low-latency 1P6D TP4-prefill / TP4-decode (6 decode workers)
+      - spec-decoding: mtp
+        conc-list: [45]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 1
+          tp: 4
+          ep: 4
+          dp-attn: true
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml:zip_override_mtp_agentx_lowlat[2]"
+        decode:
+          num-worker: 6
+          tp: 4
+          ep: 1
+          dp-attn: false
+      # high-throughput 2P1D DEP8-prefill / DEP16-decode
+      - spec-decoding: mtp
+        conc-list: [128]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 2
+          tp: 8
+          ep: 8
+          dp-attn: true
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml:zip_override_mtp_agentx_hightpt[0]"
+        decode:
+          num-worker: 1
+          tp: 16
+          ep: 16
+          dp-attn: true
+      - spec-decoding: mtp
+        conc-list: [192]
+        kv-offloading: dram
+        kv-offload-backend: { name: hicache }
+        prefill:
+          num-worker: 2
+          tp: 8
+          ep: 8
+          dp-attn: true
+          additional-settings:
+          - "CONFIG_FILE=recipes/sglang/glm5.2/gb300-fp4/agentic/glm5.2-agentx.yaml:zip_override_mtp_agentx_hightpt[1]"
+        decode:
+          num-worker: 1
+          tp: 16
+          ep: 16
+          dp-attn: true
+
 qwen3.5-fp8-gb200-dynamo-sglang-mtp:
   image: lmsysorg/sglang:v0.5.14-cu130@sha256:5027e95bf6ec536856b1b52a91d1f35ff5c564ab83e8a94758a169ff09bb8df3
   model: Qwen/Qwen3.5-397B-A17B-FP8
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index 889d4db48..da3f8d6dd 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -5972,6 +5972,17 @@
     - "Use KV-aware Dynamo routing with 4-hour correlation-ID affinity, authoritative vLLM KV events, KV-cache token metrics, and every logical vLLM server-metrics endpoint."
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2567
 
+- config-keys:
+    - glm5.2-fp4-gb300-dynamo-sglang-agentic-agg
+    - glm5.2-fp4-gb300-dynamo-sglang-agentic-disagg
+  scenario-type:
+    - agentic-coding
+  description:
+    - "Add GLM-5.2-NVFP4 GB300 AgentX (agentic-coding) Dynamo-SGLang benchmarks over a 7-point sweep: aggregated single-worker TP8 (conc 2/4), disaggregated low-latency TP4-prefill/TP4-decode 1P2D (conc 48), 1P4D (conc 48), 1P6D (conc 45), and high-throughput 2P1D DEP8-prefill/DEP16-decode (conc 128/192)"
+    - "EAGLE speculative decoding, hierarchical KV cache (hicache-size 135), MTP, decode mem-fraction-static 0.9, nixl KV transfer with UCX_TLS=ib, and X-Dynamo-Session-ID session affinity; driven from the NVIDIA/srt-slurm base+zip_override recipes via CONFIG_FILE selectors"
+    - "Image: lmsysorg/sglang:nightly-dev-cu13-20260805-211ee642; runner: gb300-nv"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2608
+
 - config-keys:
     - minimaxm3-fp4-b300-vllm-agentic-mtp
   scenario-type:
diff --git a/runners/launch_gb300-nv.sh b/runners/launch_gb300-nv.sh
index 54d208bc7..83252461d 100644
--- a/runners/launch_gb300-nv.sh
+++ b/runners/launch_gb300-nv.sh
@@ -65,6 +65,9 @@ elif [[ $MODEL_PREFIX == "glm5.1" && $PRECISION == "fp4" ]]; then
     # in our GLM-5.1 sglang recipes.
     export MODEL_PATH=/scratch/models/GLM-5.1-NVFP4
     export SRT_SLURM_MODEL_PREFIX="glm-5-fp4"
+elif [[ $MODEL_PREFIX == "glm5.2" && $PRECISION == "fp4" ]]; then
+    export MODEL_PATH=/scratch/models/GLM-5.2-NVFP4
+    export SRT_SLURM_MODEL_PREFIX="glm-5.2-fp4"
 elif [[ $MODEL_PREFIX == "glm5" && $PRECISION == "fp4" ]]; then
     export MODEL_PATH=/scratch/models/GLM-5-NVFP4
     export SRT_SLURM_MODEL_PREFIX="glm-5-fp4"
@@ -94,7 +97,7 @@ elif [[ $MODEL_PREFIX == "qwen3.5" && $PRECISION == "fp8" ]]; then
     export MODEL_PATH=/scratch/models/Qwen3.5-397B-A17B-FP8
     export SRT_SLURM_MODEL_PREFIX="qwen3.5-fp8"
 else
-    echo "Unsupported model: $MODEL_PREFIX-$PRECISION. Supported models are: dsr1-fp4, dsr1-fp8, dsv4-fp4, glm5-fp4, glm5-fp8, minimaxm2.5-fp4, minimaxm2.5-fp8, kimik2.5-fp4, qwen3.5-fp4, qwen3.5-fp8"
+    echo "Unsupported model: $MODEL_PREFIX-$PRECISION. Supported models are: dsr1-fp4, dsr1-fp8, dsv4-fp4, glm5-fp4, glm5-fp8, glm5.2-fp4, minimaxm2.5-fp4, minimaxm2.5-fp8, kimik2.5-fp4, qwen3.5-fp4, qwen3.5-fp8"
     exit 1
 fi
 
@@ -221,6 +224,16 @@ elif [[ "$IS_AGENTIC" == "1" && $FRAMEWORK == "dynamo-sglang" && $MODEL_PREFIX =
     mkdir -p recipes/sglang/deepseek-v4/agentic
     cp -rT "$GITHUB_WORKSPACE/benchmarks/multi_node/srt-slurm-recipes/sglang/deepseek-v4/agentic" \
         recipes/sglang/deepseek-v4/agentic
+elif [[ "$IS_AGENTIC" == "1" && $FRAMEWORK == "dynamo-sglang" && $MODEL_PREFIX == "glm5.2" ]]; then
+    # GLM-5.2 GB300 sglang AgentX: srt-slurm main has the agentx-mvp scenario,
+    # the zip_override sweep selectors, and the multi-frontend session-affinity
+    # schema these recipes need.
+    git clone https://github.com/NVIDIA/srt-slurm.git "$SRT_REPO_DIR"
+    cd "$SRT_REPO_DIR"
+    git checkout main
+    mkdir -p recipes/sglang/glm5.2/gb300-fp4
+    cp -rT "$GITHUB_WORKSPACE/benchmarks/multi_node/srt-slurm-recipes/sglang/glm5.2/gb300-fp4" \
+        recipes/sglang/glm5.2/gb300-fp4
 elif [[ "$IS_AGENTIC" == "1" ]]; then
     # Agentic recipes use NVIDIA/srt-slurm v1.0.36. This is the upstream
     # version validated in InferenceX PR #2302 and includes per-node DP,