fix: recover PR 2752 ingest via sweep reuse

perf-changelog.yaml

@@ -6514,3 +6514,15 @@
     - "Pin throughput runs to the committed golden thinking_on acceptance length of 2.32 at three speculative tokens; eval-only runs keep real target verification."
     - "Keep the bfloat16 Mamba SSM state the cookbook specifies: SGLang requires it on SM100 or newer whenever the flashinfer linear-attention decode backend is selected."
   pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2752
+
+- config-keys:
+    - qwen3.8next-fp4-b300-sglang-agentic-mtp
+  scenario-type:
+    - agentic-coding
+  description:
+    - "Add the day-zero Qwen3.8-Flash-Next NVFP4 AgentX recipe on B300 with SGLang native NEXTN MTP at TP4 and concurrency 1/4/8/12/16."
+    - "Serve RadixArk/Qwen3.8-Flash-Next-NVFP4 with modelopt_fp4 quantization, the trtllm_mha attention backend, and the flashinfer_trtllm MoE runner, following the Qwen3.5 NVFP4 B300 sibling."
+    - "Correct the serve flags to the SGLang cookbook's verified single-node command for this model: TP1 rather than TP4, the flashinfer linear-attention prefill and decode backends, bfloat16 Mamba SSM, and checkpoint-derived quantization."
+    - "Pin throughput runs to the committed golden thinking_on acceptance length of 2.32 at three speculative tokens; eval-only runs keep real target verification."
+    - "Keep the bfloat16 Mamba SSM state the cookbook specifies: SGLang requires it on SM100 or newer whenever the flashinfer linear-attention decode backend is selected."
+  pr-link: https://github.com/SemiAnalysisAI/InferenceX/pull/2758