fix Dockerfile

runpod/runpod-worker-oobabooga-api/Dockerfile

@@ -10,9 +10,6 @@ FROM ${DEV_IMAGE} as builder
 # DOCKER_BUILDKIT=1 docker build --build-arg MODEL_NAME="PygmalionAI/pygmalion-350m" -t magn418/runpod-oobabooga-pygmalion:test .
 # docker builder prune
 
-ARG MODEL_NAME="PygmalionAI/pygmalion-350m"
-ENV MODEL_NAME=${MODEL_NAME}
-
 WORKDIR /
 SHELL ["/bin/bash", "-o", "pipefail", "-c"]
 ENV DEBIAN_FRONTEND noninteractive\
@@ -73,23 +70,35 @@ RUN mkdir -p /workspace
 WORKDIR /workspace
 
 #RUN mkdir /workspace && 
-RUN cd /workspace && git clone https://github.com/oobabooga/text-generation-webui.git && \
-  cd /workspace/text-generation-webui && pip3 install -r requirements.txt && \
-  cd extensions/api && pip3 install -r requirements.txt
+RUN --mount=type=cache,target=/root/.cache,sharing=locked pip3 \
+    cd /workspace && git clone --depth 1 https://github.com/oobabooga/text-generation-webui.git && \
+    cd /workspace/text-generation-webui && pip3 install -r requirements.txt && \
+    cd extensions/api && pip3 install -r requirements.txt
 #  pip3 cache purge
 
-RUN cd /workspace/text-generation-webui/ && mkdir repositories && cd repositories && \
+RUN --mount=type=cache,target=/root/.cache,sharing=locked pip3 \
+    cd /workspace/text-generation-webui/ && mkdir repositories && cd repositories && \
 # https://github.com/oobabooga/GPTQ-for-LLaMa
-    git clone --branch cuda --single-branch https://github.com/qwopqwop200/GPTQ-for-LLaMa.git && \
-    (cd GPTQ-for-LLaMa && python3 setup_cuda.py bdist_wheel -d .)
+#    git clone --branch cuda --single-branch https://github.com/qwopqwop200/GPTQ-for-LLaMa.git && \
+#    (cd GPTQ-for-LLaMa && python3 setup_cuda.py bdist_wheel -d .)
+    git clone --depth 1 https://github.com/qwopqwop200/GPTQ-for-LLaMa.git && \
+    (cd GPTQ-for-LLaMa && pip3 install -r requirements.txt)
 # && python3 setup_cuda.py install
 
 
 FROM builder AS modeldownloader
-RUN cd /workspace/text-generation-webui && python3 download-model.py ${MODEL_NAME}
+
+ARG MODEL_NAME="PygmalionAI/pygmalion-350m"
+ENV MODEL_NAME=${MODEL_NAME}
+
+#RUN cd /workspace/text-generation-webui && python3 download-model.py ${MODEL_NAME}
 #RUN git lfs install && \
 #    git clone --depth 1 https://huggingface.co/${MODEL_NAME}
-
+RUN wget -P /workspace/text-generation-webui/models/ https://raw.githubusercontent.com/BlinkDL/ChatRWKV/main/v2/20B_tokenizer.json && \
+    wget -P /workspace/text-generation-webui/models/ https://huggingface.co/BlinkDL/rwkv-4-raven/resolve/main/RWKV-4-Raven-7B-v9-Eng99%25-Other1%25-20230412-ctx8192.pth
+    #https://huggingface.co/BlinkDL/rwkv-4-pile-7b/resolve/main/RWKV-4-Pile-7B-20230406-ctx8192-test949.pth
+#RUN cd /workspace/text-generation-webui && python3 download-model.py MetaIX/GPT4-X-Alpaca-30B-Int4 --text-only && \
+#    wget -P /workspace/text-generation-webui/models/GPT4-X-Alpaca-30B-Int4/ https://huggingface.co/MetaIX/GPT4-X-Alpaca-30B-Int4/resolve/main/gpt4-x-alpaca-30b-4bit.safetensors
 
 FROM ${BASE_IMAGE}
 #ENV TORCH_CUDA_ARCH_LIST="3.5;5.0;6.0;6.1;7.0;7.5;8.0;8.6+PTX"