Update Dockerfile

Dockerfile

@@ -1,28 +1,21 @@
-FROM nvidia/cuda:12.4.1-runtime-ubuntu22.04
+FROM vllm/vllm-openai:latest
 
 ENV DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
 
-# System deps
-RUN apt-get update && apt-get install -y --no-install-recommends \
-    python3.11 python3.11-venv python3-pip curl && \
-    ln -sf /usr/bin/python3.11 /usr/bin/python3 && \
-    ln -sf /usr/bin/python3.11 /usr/bin/python && \
-    apt-get clean && rm -rf /var/lib/apt/lists/*
-
-# Python packages — vLLM + TriAttention + FastAPI
+# Gemma 4 requires transformers 5.5.0+
 RUN pip install --no-cache-dir --upgrade pip && \
     pip install --no-cache-dir \
-    vllm>=0.15.0 \
-    aither-kvcache[vllm]>=2.0.0 \
+    "transformers>=5.5.0" \
     gradio>=5.0 \
     fastapi \
     uvicorn \
-    openai \
     httpx \
     requests \
-    PyMuPDF \
-    huggingface_hub
+    PyMuPDF
+
+# TriAttention (optional - fails gracefully if unavailable)
+RUN pip install --no-cache-dir aither-kvcache || true
 
 WORKDIR /app
 COPY . /app