Merge branch 'oobabooga:main' into main

README.md

@@ -212,7 +212,7 @@ Optionally, you can use the following command-line flags:
 | `--wbits WBITS`            | GPTQ: Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported. |
 | `--model_type MODEL_TYPE`  | GPTQ: Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported. |
 | `--groupsize GROUPSIZE`    | GPTQ: Group size. |
-| `--pre_layer PRE_LAYER`    | GPTQ: The number of layers to preload. |
+| `--pre_layer PRE_LAYER`    | GPTQ: The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models. |
 
 #### FlexGen
 

modules/shared.py

@@ -102,7 +102,7 @@ parser.add_argument('--threads', type=int, default=0, help='Number of threads to
 parser.add_argument('--wbits', type=int, default=0, help='GPTQ: Load a pre-quantized model with specified precision in bits. 2, 3, 4 and 8 are supported.')
 parser.add_argument('--model_type', type=str, help='GPTQ: Model type of pre-quantized model. Currently LLaMA, OPT, and GPT-J are supported.')
 parser.add_argument('--groupsize', type=int, default=-1, help='GPTQ: Group size.')
-parser.add_argument('--pre_layer', type=int, default=0, help='GPTQ: The number of layers to preload.')
+parser.add_argument('--pre_layer', type=int, default=0, help='GPTQ: The number of layers to allocate to the GPU. Setting this parameter enables CPU offloading for 4-bit models.')
 parser.add_argument('--gptq-bits', type=int, default=0, help='DEPRECATED: use --wbits instead.')
 parser.add_argument('--gptq-model-type', type=str, help='DEPRECATED: use --model_type instead.')
 parser.add_argument('--gptq-pre-layer', type=int, default=0, help='DEPRECATED: use --pre_layer instead.')

server.py

@@ -335,6 +335,7 @@ def create_interface():
                                 shared.gradio['download'] = gr.File()
                                 shared.gradio['download_button'] = gr.Button(value='Click me')
                     with gr.Tab('Upload character'):
+                        gr.Markdown("# JSON format")
                         with gr.Row():
                             with gr.Column():
                                 gr.Markdown('1. Select the JSON file')
@@ -343,10 +344,11 @@ def create_interface():
                                 gr.Markdown('2. Select your character\'s profile picture (optional)')
                                 shared.gradio['upload_img_bot'] = gr.File(type='binary', file_types=['image'])
                         shared.gradio['Upload character'] = gr.Button(value='Submit')
+
+                        gr.Markdown("# TavernAI PNG format")
+                        shared.gradio['upload_img_tavern'] = gr.File(type='binary', file_types=['image'])
                     with gr.Tab('Upload your profile picture'):
                         shared.gradio['upload_img_me'] = gr.File(type='binary', file_types=['image'])
-                    with gr.Tab('Upload TavernAI Character Card'):
-                        shared.gradio['upload_img_tavern'] = gr.File(type='binary', file_types=['image'])
 
             with gr.Tab("Parameters", elem_id="parameters"):
                 with gr.Box():