diff --git a/baseline.muh b/baseline.muh
index 3885bcc3..839af03e 100644
--- a/baseline.muh
+++ b/baseline.muh
@@ -1,4 +1,5 @@
 # baseline.muh — Competition vllm launch configuration
+# SYNCED FROM computility-run.yaml (the actual deployment config)
 #
 # This file stores ONLY the vllm server launch config.
 # Kernel tuning values live in muh/include/muh/tuning/tuning_*.cuh
@@ -7,16 +8,19 @@
 # Pipeline:
 #   muh/tuning/*.cuh (bi100_* values) → gen_patch.py → vllm kernel patches
 #   baseline.muh (vllm config)        → gen_yaml.py → computility-run.yaml
+#
+# CRITICAL: computility-run.yaml is the deployment source of truth.
+# This .muh must stay in sync with it.
 
 # --- vllm launch configuration ---
 vllm:
   model_path: /model
   served_model_name: llm
-  max_model_len: 100000
-  gpu_memory_utilization: 0.9
+  max_model_len: 256000
+  gpu_memory_utilization: 0.95
   tensor_parallel: 4
-  max_num_seqs: 1
-  max_num_batched_tokens: 8192
+  max_num_seqs: 2
+  max_num_batched_tokens: 4096
   max_seq_len_to_capture: 32768
   trust_remote_code: true
   disable_log_requests: true
@@ -26,8 +30,19 @@ vllm:
   tool_call_parser: qwen3_coder
   reasoning_parser: qwen3
   enable_prefix_caching: true
+  enforce_eager: true
+  dtype: half
 
 concurrency: 1
 
 env:
   VLLM_ENGINE_ITERATION_TIMEOUT_S: 3600
+  VLLM_ATTENTION_BACKEND: XFORMERS
+  ENABLE_CUSTOM_IPC: 1
+  PYTHONPATH: /usr/local/corex/lib/python3/dist-packages:/usr/local/corex/lib64/python3/dist-packages
+  LD_LIBRARY_PATH: /usr/local/corex/lib64:/usr/local/openmpi/lib
+  VLLM_COREX_FA2_LIBRARY: /usr/local/corex/lib64/libcorex_fa2.so
+  VLLM_COREX_GDN_LIBRARY: /usr/local/corex/lib64/libcorex_gdn.so
+  VLLM_COREX_MOE_LIBRARY: /usr/local/corex/lib64/libcorex_moe.so
+  VLLM_REQUEST_METRICS_FILE: /tmp/vllm-request-metrics.jsonl
+  VLLM_CACHE_BLOCK_SIZE: 16