Bllossom
/

llama-3.1-Korean-Bllossom-Vision-8B

Image-Text-to-Text

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

ShinDJ commited on Sep 4

Commit

0e37ea4

•

1 Parent(s): 826088a

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -94,7 +94,7 @@ inputs = processor(
     chat_messages,
     image,
     return_tensors='pt',
-).to('cuda:0')
 output = model.generate(
     **inputs,
@@ -134,7 +134,7 @@ with torch.no_grad():
     )
     bos_token = processor.tokenizer.bos_token_id
-    chat_messages = torch.cat([torch.tensor([[bos_token]]),chat_messages],dim=-1).to('cuda:0')
     output = model.generate(

     chat_messages,
     image,
     return_tensors='pt',
+)
 output = model.generate(
     **inputs,
     )
     bos_token = processor.tokenizer.bos_token_id
+    chat_messages = torch.cat([torch.tensor([[bos_token]]),chat_messages],dim=-1)
     output = model.generate(