make server run

localagi · May 24, 2023 · 579b02c · 579b02c
1 parent 817e4bb
commit 579b02c
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 4 deletions.
diff --git a/alias.llama.cpp b/alias.llama.cpp
@@ -1,4 +1,4 @@
 # Alias helper
 source .env
 GPU_ARGS="--gpus all"
-alias llama.cpp="docker run ${GPU_ARGS} -v ${LOCAL_MODEL_DIR}:${LOCAL_MODEL_DIR} localagi/llama.cpp:${LLAMA_CPP_VERSION}${LLAMA_CPP_FLAVOR}"
+alias llama.cpp="docker run --init ${GPU_ARGS} -v ${LOCAL_MODEL_DIR}:${LOCAL_MODEL_DIR} localagi/llama.cpp:${LLAMA_CPP_VERSION}${LLAMA_CPP_FLAVOR}"
diff --git a/docker-compose.yaml b/docker-compose.yaml
@@ -4,9 +4,9 @@ services:
 # GPU Inference        
   llama.cpp-server:
     image: localagi/llama.cpp:${LLAMA_CPP_VERSION}${LLAMA_CPP_FLAVOR}
-    command: ["--model", "/models/Wizard-Vicuna-13B-Uncensored-GGML/Wizard-Vicuna-13B-Uncensored.ggml.q5_1.bin" ]
-    init: true
-    tty: true
+    command: "server --model ${LOCAL_MODEL_DIR}/Wizard-Vicuna-7B-Uncensored-GGML/Wizard-Vicuna-7B-Uncensored.ggmlv3.q5_1.bin --host 0.0.0.0"
+    ports:
+      - 8080:8080
     volumes:
       - $LOCAL_MODEL_DIR:$LOCAL_MODEL_DIR
     deploy: