Add an option to put all quantization variants in the same repo

app.py

@@ -87,7 +87,7 @@ def split_upload_model(model_path, repo_id, oauth_token: gr.OAuthToken | None, s
     
     print("Sharded model has been uploaded successfully!")
 
-def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_repo, train_data_file, split_model, split_max_tensors, split_max_size, oauth_token: gr.OAuthToken | None):
+def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_repo, same_repo, train_data_file, split_model, split_max_tensors, split_max_size, oauth_token: gr.OAuthToken | None):
     if oauth_token.token is None:
         raise ValueError("You must be logged in to use GGUF-my-repo")
     model_name = model_id.split('/')[-1]
@@ -156,7 +156,10 @@ def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_rep
         print(f"Quantized model path: {quantized_gguf_path}")
 
         # Create empty repo
-        new_repo_url = api.create_repo(repo_id=f"{username}/{model_name}-{imatrix_q_method if use_imatrix else q_method}-GGUF", exist_ok=True, private=private_repo)
+        q_suffix = imatrix_q_method if use_imatrix else q_method
+        repo_name = f"{username}/{model_name}-GGUF" if same_repo else f"{username}/{model_name}-{q_suffix}-GGUF"
+        
+        new_repo_url = api.create_repo(repo_id=repo_name, exist_ok=True, private=private_repo)
         new_repo_id = new_repo_url.repo_id
         print("Repo created successfully!", new_repo_url)
 
@@ -305,6 +308,12 @@ with gr.Blocks(css=css) as demo:
         info="Create a private repo under your username."
     )
 
+    same_repo = gr.Checkbox(
+        value=False,
+        label="Same Repo",
+        info="Put all quantization variants in the same repo."
+    )
+
     train_data_file = gr.File(
         label="Training Data File",
         file_types=["txt"],
@@ -347,6 +356,7 @@ with gr.Blocks(css=css) as demo:
             use_imatrix,
             imatrix_q_method,
             private_repo,
+            same_repo,
             train_data_file,
             split_model,
             split_max_tensors,