Update Dockerfile

Dockerfile

@@ -1,35 +1,55 @@
-FROM python:3.10-slim
+# STAGE 1: The Builder (Heavy Lifting)
+FROM python:3.10-slim AS builder
+
+WORKDIR /build
 
-# Install Rust + build tools
+# Install build dependencies
 RUN apt-get update && apt-get install -y --no-install-recommends \
     build-essential curl git rustc cargo \
     && rm -rf /var/lib/apt/lists/*
 
-WORKDIR /app
-
-# 1. Clone rustbpe and build it
+# Build Karpathy's rustbpe
 RUN git clone https://github.com/karpathy/rustbpe.git \
     && pip install --no-cache-dir maturin \
-    && cd rustbpe && maturin build --release --out dist \
-    && pip install dist/*.whl
+    && cd rustbpe && maturin build --release --out dist
+
+# STAGE 2: The Final App (Slim & Fast)
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install only runtime dependencies
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    libstdc++6 \
+    && rm -rf /var/lib/apt/lists/*
 
-# 2. Install requirements
+# Copy the wheel from the builder and install it
+COPY --from=builder /build/rustbpe/dist/*.whl .
+RUN pip install *.whl && rm *.whl
+
+# Install requirements (Optimized for CPU/Space)
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt \
     && pip install --no-cache-dir torch --index-url https://download.pytorch.org/whl/cpu
 
-# 3. Copy your project files
+# Copy project files
 COPY . .
 
-# --- THE PRO FIX START ---
-# Create the specific directory the library is hunting for
-# and move the uploaded tokenizer.pkl into it.
-RUN mkdir -p /root/.cache/nanochat/tokenizer/ \
-    && mv tokenizer.pkl /root/.cache/nanochat/tokenizer/tokenizer.pkl
-# --- THE PRO FIX END ---
+# --- THE PRO FIX (Correcting the paths) ---
+# Note: Hugging Face Spaces run as user "user" (UID 1000), not root.
+# We create the cache for both potential users to be safe.
+RUN mkdir -p /root/.cache/nanochat/tokenizer/ && \
+    cp tokenizer.pkl /root/.cache/nanochat/tokenizer/tokenizer.pkl && \
+    cp token_bytes.pt /root/.cache/nanochat/tokenizer/token_bytes.pt
 
-RUN pip cache purge
+# Set permissions for the HF "user"
+RUN mkdir -p /.cache/nanochat/tokenizer/ && \
+    chmod -R 777 /.cache && \
+    cp tokenizer.pkl /.cache/nanochat/tokenizer/tokenizer.pkl && \
+    cp token_bytes.pt /.cache/nanochat/tokenizer/token_bytes.pt
 
+# Clean up and finish
+RUN pip cache purge
 EXPOSE 7860
 ENV GRADIO_SERVER_NAME="0.0.0.0"