Create Dockerfile

Dockerfile

@@ -0,0 +1,10 @@
+FROM ghcr.io/huggingface/text-generation-inference:latest
+
+# Define the model to use
+ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
+
+# Set the number of GPU shards (adjust based on GPU availability)
+ENV NUM_SHARD=1
+
+# Run the TGI server
+CMD ["--model-id", "${MODEL_ID}", "--port", "8080", "--num-shard", "${NUM_SHARD}"]