adaptiveaiventures commited on
Commit
f48b421
·
verified ·
1 Parent(s): c9ad6e4

Create Dockerfile

Browse files
Files changed (1) hide show
  1. Dockerfile +10 -0
Dockerfile ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ FROM ghcr.io/huggingface/text-generation-inference:latest
2
+
3
+ # Define the model to use
4
+ ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
5
+
6
+ # Set the number of GPU shards (adjust based on GPU availability)
7
+ ENV NUM_SHARD=1
8
+
9
+ # Run the TGI server
10
+ CMD ["--model-id", "${MODEL_ID}", "--port", "8080", "--num-shard", "${NUM_SHARD}"]