diff --git a/.github/configs/nvidia-master.yaml b/.github/configs/nvidia-master.yaml
index 5b550879c..47d66aa2d 100644
--- a/.github/configs/nvidia-master.yaml
+++ b/.github/configs/nvidia-master.yaml
@@ -1669,6 +1669,29 @@ dsr1-fp4-b200-sglang:
     - { tp: 4, ep: 4, conc-start: 4, conc-end: 128 }
     - { tp: 8, ep: 8, conc-start: 4, conc-end: 16 }
 
+# NOTE: At the time of submission, https://cookbook.sglang.io/autoregressive/DeepSeek/DeepSeek-R1
+# does not have a B300-specific recipe, so this config reuses the existing DSR1 FP4
+# B200 SGLang recipe as-is until B300-specific tuning is available.
+dsr1-fp4-b300-sglang:
+  image: lmsysorg/sglang:v0.5.10.post1-cu130
+  model: nvidia/DeepSeek-R1-0528-FP4-V2
+  model-prefix: dsr1
+  runner: b300
+  precision: fp4
+  framework: sglang
+  multinode: false
+  seq-len-configs:
+  - isl: 1024
+    osl: 1024
+    search-space:
+    - { tp: 4, ep: 4, conc-start: 4, conc-end: 128 }
+    - { tp: 8, ep: 8, conc-start: 4, conc-end: 128 }
+  - isl: 8192
+    osl: 1024
+    search-space:
+    - { tp: 4, ep: 4, conc-start: 4, conc-end: 128 }
+    - { tp: 8, ep: 8, conc-start: 4, conc-end: 16 }
+
 dsr1-fp4-b200-trt:
   image: nvcr.io#nvidia/tensorrt-llm/release:1.2.0rc6.post2
   model: nvidia/DeepSeek-R1-0528-FP4-V2
diff --git a/benchmarks/single_node/dsr1_fp4_b300.sh b/benchmarks/single_node/dsr1_fp4_b300.sh
new file mode 100644
index 000000000..7c26f3cba
--- /dev/null
+++ b/benchmarks/single_node/dsr1_fp4_b300.sh
@@ -0,0 +1,81 @@
+#!/usr/bin/env bash
+
+# NOTE: At the time of submission, https://cookbook.sglang.io/autoregressive/DeepSeek/DeepSeek-R1
+# does not have a B300-specific recipe, so this script reuses the existing
+# DSR1 FP4 B200 SGLang recipe as-is until B300-specific tuning is available.
+
+source "$(dirname "$0")/../benchmark_lib.sh"
+
+check_env_vars \
+    MODEL \
+    TP \
+    CONC \
+    ISL \
+    OSL \
+    RANDOM_RANGE_RATIO \
+    RESULT_FILENAME \
+    EP_SIZE
+
+if [[ -n "$SLURM_JOB_ID" ]]; then
+  echo "JOB $SLURM_JOB_ID running on $SLURMD_NODENAME"
+fi
+
+hf download "$MODEL"
+
+nvidia-smi
+
+SERVER_LOG=/workspace/server.log
+PORT=${PORT:-8888}
+
+# Default: recv every ~10 requests; if CONC ≥ 16, relax to ~30 requests between scheduler recv polls.
+if [[ $CONC -ge 16 ]]; then
+  SCHEDULER_RECV_INTERVAL=30
+else
+  SCHEDULER_RECV_INTERVAL=10
+fi
+echo "SCHEDULER_RECV_INTERVAL: $SCHEDULER_RECV_INTERVAL, CONC: $CONC, ISL: $ISL, OSL: $OSL"
+
+EVAL_CONTEXT_ARGS=""
+if [ "${EVAL_ONLY}" = "true" ]; then
+    setup_eval_context
+    EVAL_CONTEXT_ARGS="--context-length $EVAL_MAX_MODEL_LEN"
+fi
+# Start GPU monitoring (power, temperature, clocks every second)
+start_gpu_monitor
+
+set -x
+PYTHONNOUSERSITE=1 python3 -m sglang.launch_server --model-path $MODEL --host 0.0.0.0 --port $PORT --trust-remote-code \
+--tensor-parallel-size=$TP --data-parallel-size=1 \
+--cuda-graph-max-bs 256 --max-running-requests 256 --mem-fraction-static 0.85 --kv-cache-dtype fp8_e4m3 \
+--chunked-prefill-size 16384 \
+--ep-size $EP_SIZE --quantization modelopt_fp4 --enable-flashinfer-allreduce-fusion --scheduler-recv-interval $SCHEDULER_RECV_INTERVAL \
+--enable-symm-mem --disable-radix-cache --attention-backend trtllm_mla --moe-runner-backend flashinfer_trtllm --stream-interval 10 $EVAL_CONTEXT_ARGS > $SERVER_LOG 2>&1 &
+
+SERVER_PID=$!
+
+# Wait for server to be ready
+wait_for_server_ready --port "$PORT" --server-log "$SERVER_LOG" --server-pid "$SERVER_PID"
+
+pip install -q datasets pandas
+
+run_benchmark_serving \
+    --model "$MODEL" \
+    --port "$PORT" \
+    --backend vllm \
+    --input-len "$ISL" \
+    --output-len "$OSL" \
+    --random-range-ratio "$RANDOM_RANGE_RATIO" \
+    --num-prompts $((CONC * 10)) \
+    --max-concurrency "$CONC" \
+    --result-filename "$RESULT_FILENAME" \
+    --result-dir /workspace/
+
+# After throughput, run evaluation only if RUN_EVAL is true
+if [ "${RUN_EVAL}" = "true" ]; then
+    run_eval --framework lm-eval --port "$PORT"
+    append_lm_eval_summary
+fi
+
+# Stop GPU monitoring
+stop_gpu_monitor
+set +x
diff --git a/perf-changelog.yaml b/perf-changelog.yaml
index 3f081c240..75fae41e3 100644
--- a/perf-changelog.yaml
+++ b/perf-changelog.yaml
@@ -1396,3 +1396,11 @@
     - "Image: lmsysorg/sglang:v0.5.10.post1-cu130"
     - "TP=4, concurrency 4-256 for 1k1k and 8k1k"
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1048
+
+- config-keys:
+    - dsr1-fp4-b300-sglang
+  description:
+    - "Add DeepSeek-R1-0528 FP4 B300 SGLang benchmark (non-MTP)"
+    - "Image: lmsysorg/sglang:v0.5.10.post1-cu130"
+    - "At the time of submission, https://cookbook.sglang.io/autoregressive/DeepSeek/DeepSeek-R1 does not have a B300-specific recipe, so this reuses the existing DSR1 FP4 B200 SGLang recipe as-is"
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/1049
diff --git a/runners/launch_b300-nv.sh b/runners/launch_b300-nv.sh
index 3d863b54c..7c042b4f1 100644
--- a/runners/launch_b300-nv.sh
+++ b/runners/launch_b300-nv.sh
@@ -217,7 +217,12 @@ find . -name '.nfs*' -delete 2>/dev/null || true
 else
 
     HF_HUB_CACHE_MOUNT="/scratch/models"
-    export MODEL="/scratch/models/${MODEL#*/}"
+    # Qwen3.5-397B-A17B-FP8 is pre-staged under /scratch/models on the B300 cluster,
+    # so point MODEL at the local copy. Other models fall through and use `hf download`
+    # against the mounted cache from their benchmark script.
+    if [[ "$MODEL" == "Qwen/Qwen3.5-397B-A17B-FP8" ]]; then
+        export MODEL="/scratch/models/${MODEL#*/}"
+    fi
     SQUASH_FILE="/data/squash/$(echo "$IMAGE" | sed 's/[\/:@#]/_/g').sqsh"
     FRAMEWORK_SUFFIX=$([[ "$FRAMEWORK" == "trt" ]] && printf '_trt' || printf '')
     SPEC_SUFFIX=$([[ "$SPEC_DECODING" == "mtp" ]] && printf '_mtp' || printf '')