diff --git a/renderers/base.py b/renderers/base.py
index f3418841..582e102b 100644
--- a/renderers/base.py
+++ b/renderers/base.py
@@ -988,7 +988,8 @@ def is_multimodal(r: object) -> bool:
     "Qwen/Qwen3.5-35B-A3B": "qwen3.5",
     "Qwen/Qwen3.5-122B-A10B": "qwen3.5",
     "Qwen/Qwen3.5-397B-A17B": "qwen3.5",
-    # Qwen3.6.
+    # Qwen3.6. Both sizes ship the same chat template.
+    "Qwen/Qwen3.6-27B": "qwen3.6",
     "Qwen/Qwen3.6-35B-A3B": "qwen3.6",
     # Qwen3.8.
     "Qwen/Qwen3.8-27B": "qwen3.8",
@@ -1108,6 +1109,7 @@ def is_multimodal(r: object) -> bool:
     "Qwen/Qwen3.5-397B-A17B": {"image"},
     # Qwen3.6 extends Qwen3.5's chat template; same VL bits, only
     # tool-call argument serialization differs.
+    "Qwen/Qwen3.6-27B": {"image"},
     "Qwen/Qwen3.6-35B-A3B": {"image"},
     # Qwen3.8 adds reasoning-effort control and preserves thinking by default.
     "Qwen/Qwen3.8-27B": {"image"},
diff --git a/renderers/qwen35.py b/renderers/qwen35.py
index 8f2dbffb..a737fcc4 100644
--- a/renderers/qwen35.py
+++ b/renderers/qwen35.py
@@ -105,6 +105,7 @@
     "Qwen/Qwen3.5-122B-A10B": True,
     "Qwen/Qwen3.5-397B-A17B": True,
     # Qwen3.6 extends the Qwen3.5 template; same big-size polarity.
+    "Qwen/Qwen3.6-27B": True,
     "Qwen/Qwen3.6-35B-A3B": True,
     # Qwen3.8 keeps thinking enabled by default.
     "Qwen/Qwen3.8-27B": True,
diff --git a/tests/parity.py b/tests/parity.py
index ac34e11c..e0fb94a3 100644
--- a/tests/parity.py
+++ b/tests/parity.py
@@ -109,6 +109,7 @@ def _model(
     _model("Qwen/Qwen3.5-35B-A3B", shared=False, roundtrip=False),
     _model("Qwen/Qwen3.5-122B-A10B", shared=False, roundtrip=False),
     _model("Qwen/Qwen3.5-397B-A17B", shared=False, roundtrip=False),
+    _model("Qwen/Qwen3.6-27B", shared=False, roundtrip=False),
     _model(
         "Qwen/Qwen3.6-35B-A3B",
         bridge=True,