diff --git a/k8s/apps/llm-serving/reasoning.yaml b/k8s/apps/llm-serving/reasoning.yaml
index 66dfb0e..0ae0f8c 100644
--- a/k8s/apps/llm-serving/reasoning.yaml
+++ b/k8s/apps/llm-serving/reasoning.yaml
@@ -26,14 +26,18 @@ spec:
       - --reasoning-parser=deepseek_r1
       # Without these, any client sending tool_choice="auto" (pi does, for
       # Read/Bash/etc.) gets a 400: "auto" tool choice requires
-      # --enable-auto-tool-choice and --tool-call-parser to be set. Verified
-      # against this exact deployed vLLM v0.11.0's registered parsers
-      # (ToolParserManager.tool_parsers) -- deepseek_v3 matches, same family
-      # as the deepseek_r1 reasoning-parser already set above (this model
-      # emits DeepSeek's own special-token format, not Qwen's Hermes-style,
-      # despite being a Qwen-base distillation).
+      # --enable-auto-tool-choice and --tool-call-parser to be set.
+      # deepseek_v3 (matching --reasoning-parser above) 400s here --
+      # "DeepSeek-V3 Tool parser could not locate tool call start/end
+      # tokens in the tokenizer" -- this checkpoint is a Qwen2.5-32B base
+      # distilled on R1 traces, so its tokenizer never got DeepSeek-V3's
+      # own special tool-call tokens registered even though it kept R1's
+      # <think> reasoning format. hermes parses tool calls from plain text
+      # patterns instead of special tokens, so it works against the
+      # underlying Qwen tokenizer regardless. Verified live: deepseek_v3
+      # 400s, hermes returns a real tool_calls response.
       - --enable-auto-tool-choice
-      - --tool-call-parser=deepseek_v3
+      - --tool-call-parser=hermes
       - --host=0.0.0.0
       - --port=8080
       env: