Update Dockerfile

Dockerfile

@@ -11,37 +11,33 @@ ENV OMP_NUM_THREADS=1 \
     HF_DATASETS_CACHE=/home/user/.cache/huggingface/datasets \
     OFFLOAD_DIR=/home/user/app/offload \
     MERGED_MODEL_DIR=/home/user/app/merged-model \
-    QUANTIZE=4bit \
+    QUANTIZE=none \                  # <── default: no quantization
     USE_ADAPTER_INFERENCE=1 \
     FORCE_REMERGE=0 \
     ALLOW_RUNTIME_MERGE=0 \
     NVIDIA_VISIBLE_DEVICES=all \
     NVIDIA_DRIVER_CAPABILITIES=compute,utility
 
-# Base OS deps
+# System deps
 RUN apt-get update && apt-get install -y --no-install-recommends \
     python3 python3-pip git git-lfs ca-certificates && \
     rm -rf /var/lib/apt/lists/* && \
     git lfs install
 
-# Modern torch stack aligned to cu121 and Transformers 4.44.x
+# Torch stack aligned with transformers 4.44
 RUN python3 -m pip install --upgrade pip
 RUN python3 -m pip install --no-cache-dir \
     --index-url https://download.pytorch.org/whl/cu121 \
     torch==2.3.1 torchvision==0.18.1 torchaudio==2.3.1
 
-# NOTE: Do NOT install faiss-gpu. We keep CPU FAISS only (from requirements.txt).
-# (Removing: python3 -m pip install faiss-gpu==1.7.2)
-
-# App requirements (single source of truth)
+# Only CPU FAISS (not GPU)
 COPY requirements.txt .
 RUN python3 -m pip install --no-cache-dir -r requirements.txt
 
-# App code
 COPY . .
 
-# Needed dirs
 RUN mkdir -p "$OFFLOAD_DIR" "$MERGED_MODEL_DIR" "$HF_HOME" "$TRANSFORMERS_CACHE"
 
 EXPOSE 7860
 CMD ["python3", "app.py"]
+