change port export

Dockerfile

@@ -18,9 +18,10 @@ RUN git clone https://github.com/ggerganov/llama.cpp.git llamacpp --depth 1 && \
     cd llamacpp && \
     cmake -B build -DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS && \
     cmake --build build --config Release --target main server && \
-    cp build/bin/* ~/ && \
+    cp build/bin/* ../ && \
     touch /llama.log && \
-    chmod 777 /llama.log
+    chmod 777 /llama.log && \
+    ls -lt 
 
 # Download model
 RUN mkdir model && \

README.md

@@ -1,20 +1,20 @@
 ---
-title: OpenHermes-2.5-Mistral-7B-GGUF (Q4_K_M)
+title: openchat-3.5-0106-GGUF (Q4_K_M)
 colorFrom: purple
 colorTo: blue
 sdk: docker
 models:
   - teknium/OpenHermes-2.5-Mistral-7B
-  - TheBloke/OpenHermes-2.5-Mistral-7B-GGUF
+  - TheBloke/openchat-3.5-0106-GGUF
 tags:
   - inference api
   - openai-api compatible
   - llama-cpp-python
-  - OpenHermes-2.5-Mistral-7B-GGUF
+  - openchat-3.5-0106-GGUF
   - gguf
 pinned: false
 ---
 
-# OpenHermes-2.5-Mistral-7B-GGUF (Q4_K_M)
+# openchat-3.5-0106-GGUF (Q4_K_M)
 
 Please refer to the [index.html](index.html) for more information.

start_server.sh

@@ -3,7 +3,7 @@
 # For mlock support
 ulimit -l unlimited
 
-
+/main -m model/gguf-model.bin --color -p '你好，你是谁'
 
 /server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin 
 /llamacpp/build/bin/server --host 0.0.0.0 --port 7860 -m model/gguf-model.bin