Remove guided decoding backend override and add airportsdata dependency

app.py

@@ -27,7 +27,6 @@ VLLM_ARGS = [
     "--gpu-memory-utilization", "0.90",
     "--trust-remote-code",
     "--disable-log-requests",                # reduce log noise
-    "--guided-decoding-backend", "lm-format-enforcer",  # avoid outlines issues
 ]
 if "AWQ" in MODEL_ID.upper():
     VLLM_ARGS += ["--quantization", "awq_marlin"]  # faster AWQ kernel if available

requirements.txt

@@ -9,3 +9,6 @@ vllm==0.6.3.post1
 torch==2.4.0
 transformers>=4.44
 accelerate>=0.30
+
+# Required for vLLM guided decoding (even if not actively used)
+airportsdata>=20240400