fix(qwen3.6-27b): use --default-chat-template-kwargs (not --chat-template-kwargs)

Follow-up to 29d17ed which used the wrong vLLM flag name (`--chat-template-kwargs`), causing boot failure: "vllm: error: unrecognized arguments: --chat-template-kwargs {"enable_thinking": false}". vLLM's actual flag for setting server-side default chat template kwargs is `--default-chat-template-kwargs` (with the `default-` prefix). Confirmed by - vLLM nightly source: vllm/engine/arg_utils.py defines `default_chat_template_kwargs: dict[str, Any] | None = None` with json.loads parsing. - vLLM PR #37739 ("Fix default_chat_template_kwargs handling in Responses API") references it as already available in the shared render stack. Behavior unchanged: thinking OFF by default for all 17 vLLM Qwen 27B composes (plus bounded-thinking unaffected — it intentionally keeps thinking ON). Per-request override still works via OpenAI extra_body: {"chat_template_kwargs": {"enable_thinking": true}} Verified: vllm-qwen36-27b-dual now boots cleanly with the corrected flag. Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
2026-05-10 17:44:50 +00:00
parent 29d17ed82d
commit 534d29f1b1
17 changed files with 17 additions and 17 deletions
@@ -127,7 +127,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -118,7 +118,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -141,7 +141,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -130,7 +130,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -154,7 +154,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -149,7 +149,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -257,7 +257,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -139,7 +139,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -237,7 +237,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -134,7 +134,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -138,7 +138,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -247,7 +247,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -357,7 +357,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -374,7 +374,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -274,7 +274,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -97,7 +97,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser
@@ -157,7 +157,7 @@ services:
      - --trust-remote-code
      - --reasoning-parser
      - qwen3
-      - --chat-template-kwargs
+      - --default-chat-template-kwargs
      - '{"enable_thinking": false}'
      - --enable-auto-tool-choice
      - --tool-call-parser