Add OMP_NUM_THREADS env var and improve Outlines requirement comment

- Set OMP_NUM_THREADS=1 to fix libgomp warning
- Add comment about Python 3.10+ requirement for Outlines

Dockerfile

@@ -14,7 +14,8 @@ ENV PYTHONDONTWRITEBYTECODE=1 \
     V4_ENABLE_QUANTIZATION=true \
     V4_USE_FP16_FOR_SPEED=true \
     HF_HOME=/tmp/huggingface \
-    TRANSFORMERS_NO_TORCHAO=1
+    TRANSFORMERS_NO_TORCHAO=1 \
+    OMP_NUM_THREADS=1
 
 # Set work directory
 WORKDIR /app

requirements.txt

@@ -20,7 +20,7 @@ accelerate>=0.33.0,<1.0.0  # Required for GPU quantization (V4)
 bitsandbytes>=0.44.0  # 4-bit NF4 quantization for GPU (V4)
 einops>=0.6.0,<1.0.0  # Required for model architecture (V4)
 scipy>=1.10.0,<2.0.0  # Often needed for unquantized models (V4)
-outlines>=0.0.34  # JSON schema enforcement for V4 structured summarization
+outlines>=0.0.34  # JSON schema enforcement for V4 structured summarization (requires Python 3.10+)
 
 # Testing
 pytest>=7.0.0,<8.0.0