adaptiveaiventures commited on
Commit
ba00a28
·
verified ·
1 Parent(s): b2c1ee4

Update Dockerfile

Browse files
Files changed (1) hide show
  1. Dockerfile +5 -0
Dockerfile CHANGED
@@ -3,5 +3,10 @@ FROM ghcr.io/huggingface/text-generation-inference:latest
3
  # Define Model
4
  ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
5
 
 
 
 
 
 
6
  # Run the TGI server
7
  CMD ["--model-id", "adaptiveaiventures/Llama-2-7b-chat-finetune", "--port", "8080", "--num-shard", "1", "--dtype", "bfloat16", "--max-batch-prefill-tokens", "1024", "--disable-custom-kernels"]
 
3
  # Define Model
4
  ENV MODEL_ID="adaptiveaiventures/Llama-2-7b-chat-finetune"
5
 
6
+ # Set cache directory to a writable location
7
+ ENV HF_HOME="/tmp"
8
+ ENV TRANSFORMERS_CACHE="/tmp"
9
+ ENV HUGGINGFACE_HUB_CACHE="/tmp"
10
+
11
  # Run the TGI server
12
  CMD ["--model-id", "adaptiveaiventures/Llama-2-7b-chat-finetune", "--port", "8080", "--num-shard", "1", "--dtype", "bfloat16", "--max-batch-prefill-tokens", "1024", "--disable-custom-kernels"]