Create DockerFile
Browse files- DockerFile +16 -0
DockerFile
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FROM ghcr.io/ggerganov/llama.cpp:full
|
| 2 |
+
|
| 3 |
+
# Install wget
|
| 4 |
+
RUN apt update && apt install wget -y
|
| 5 |
+
|
| 6 |
+
# Download the GGUF model file
|
| 7 |
+
RUN wget "https://huggingface.co/ggml-org/SmolVLM-256M-Instruct-GGUF/resolve/main/SmolVLM-256M-Instruct-Q8_0.gguf" -O /smoll.gguf
|
| 8 |
+
|
| 9 |
+
# Download the mmproj file
|
| 10 |
+
RUN wget "https://huggingface.co/ggml-org/SmolVLM-256M-Instruct-GGUF/resolve/main/mmproj-SmolVLM-256M-Instruct-Q8_0.gguf" -O /mmproj.gguf
|
| 11 |
+
|
| 12 |
+
# Expose port expected by Hugging Face Spaces
|
| 13 |
+
EXPOSE 7860
|
| 14 |
+
|
| 15 |
+
# Default command to start the model server
|
| 16 |
+
CMD ["--server", "-m", "/smoll.gguf", "--mmproj", "/mmproj.gguf", "--port", "7860", "--host", "0.0.0.0", "-n", "512"]
|