Upload 3 files

Dockerfile

@@ -2,8 +2,18 @@ FROM ghcr.io/huggingface/chat-ui:latest AS base
 
 FROM ghcr.io/huggingface/text-generation-inference:latest AS final
 
-ARG MODEL_NAME
-ENV MODEL_NAME=${MODEL_NAME}
+# Define arguments for multiple models
+ARG MODEL_NAME_1
+ENV MODEL_NAME_1=${BronioInt/Lake-1}
+
+ARG MODEL_NAME_2
+ENV MODEL_NAME_2=${BronioInt/Lake-1_Plus}
+
+ARG MODEL_NAME_3
+ENV MODEL_NAME_3=${BronioInt/Lake-2_Flash}
+
+ARG MODEL_NAME_3
+ENV MODEL_NAME_3=${BronioInt/Lake-1_Pro}
 
 ENV TZ=Europe/Paris \
     PORT=3000
@@ -43,7 +53,6 @@ ENV HOME=/home/user \
 RUN npm config set prefix /home/user/.local
 RUN npm install -g dotenv-cli
 
-
 # copy chat-ui from base image
 COPY --from=base --chown=1000 /app/node_modules /app/node_modules
 COPY --from=base --chown=1000 /app/package.json /app/package.json

entrypoint.sh

@@ -2,19 +2,25 @@
 
 # Make sure `/data/db` directory exists even with persistent storage
 mkdir -p /data/db
+
 # If app crashed, mongo didn't stop gracefully. Remove all the old *.lock files
-find /data/db -name "*.lock" -type f -exec rm -f {} \; 
+find /data/db -name "*.lock" -type f -exec rm -f {} \;
+
 # Start the local Mongo database
 mongod &
 
-# Start the text-generation-inference process
-text-generation-launcher --model-id ${MODEL_NAME} --num-shard 1 --port 8080 --trust-remote-code &
+# Start the text-generation-inference process for each model
+for model in "$MODEL_NAME_1" "$MODEL_NAME_2" "$MODEL_NAME_3" "$MODEL_NAME_4" "$MODEL_NAME_5"; do
+    echo "Starting text-generation-inference for model: $model"
+    text-generation-launcher --model-id "$model" --num-shard 1 --port 8080 --trust-remote-code &
+done
 
-# Wait for text-generation-inference to start
+# Wait for the text-generation-inference processes to start
 curl --retry 60 --retry-delay 10 --retry-connrefused http://127.0.0.1:8080/health
 
 # Start the chat-ui process
 dotenv -e /app/.env -c -- node /app/build/index.js -- --host 0.0.0.0 --port 3000
+
 # Wait for any process to exit
 wait -n