Spaces:

FoundationVision
/

LlamaGen

Runtime error

ShoufaChen commited on Jun 8

Commit

b7f8141

•

1 Parent(s): e61c6be

low memory

Files changed (1) hide show

app.py CHANGED Viewed

@@ -44,7 +44,7 @@ def load_model(args):
     llm = LLM(
         args=args,
         model='serve/fake_json/{}.json'.format(args.gpt_model),
-        gpu_memory_utilization=0.6,
         skip_tokenizer_init=True)
     print(f"gpt model is loaded")
     return vq_model, llm, image_size

     llm = LLM(
         args=args,
         model='serve/fake_json/{}.json'.format(args.gpt_model),
+        gpu_memory_utilization=0.5,
         skip_tokenizer_init=True)
     print(f"gpt model is loaded")
     return vq_model, llm, image_size