diff --git a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
index 65374f0..7f58b97 100644
--- a/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
+++ b/infrastructure/gpus/base/vllm-servers/rtx6000-vllm.yaml
@@ -40,7 +40,7 @@ spec:
             - --tokenizer-mode=hf 
             - --gpu-memory-utilization=0.95
             - --kv-cache-dtype=fp8_e5m2
-            - --max-model-len=131072 
+            - --max-model-len=256389
             - --enable-auto-tool-choice 
             - --enable-chunked-prefill
             - --max-num-batched-tokens=8192
@@ -48,7 +48,7 @@ spec:
             - --trust-remote-code
             - --dtype=auto 
             - --enable-prefix-caching
-            - --tool-call-parser=qwen3_xml
+            - --tool-call-parser=qwen3_coder 
             - --reasoning-parser=qwen3
             - --disable-custom-all-reduce
           ports: