diff --git "a/gradio_web_server.log" "b/gradio_web_server.log"
--- "a/gradio_web_server.log"
+++ "b/gradio_web_server.log"
@@ -1002,3 +1002,357 @@
 2024-07-10 07:59:35 | ERROR | stderr | AttributeError: module 'gradio' has no attribute 'MultimodalTextbox'
 2024-07-10 07:59:36 | INFO | stdout | IMPORTANT: You are using gradio version 4.16.0, however version 4.29.0 is available, please upgrade.
 2024-07-10 07:59:36 | INFO | stdout | --------
+2024-07-10 08:07:10 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:07:10 | INFO | stdout | 
+2024-07-10 08:07:10 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:07:21 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:07:21 | INFO | stdout | skip for now
+2024-07-10 08:07:21 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:07:21 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:07:21 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:07:21 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1514, in call_function
+2024-07-10 08:07:21 | ERROR | stderr |     prediction = await anyio.to_thread.run_sync(
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:07:21 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:07:21 | ERROR | stderr |     return await future
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:07:21 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 833, in wrapper
+2024-07-10 08:07:21 | ERROR | stderr |     response = f(*args, **kwargs)
+2024-07-10 08:07:21 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_vision_named.py", line 235, in add_text
+2024-07-10 08:07:21 | ERROR | stderr |     + [x.to_gradio_chatbot() for x in states]
+2024-07-10 08:07:21 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_vision_named.py", line 235, in <listcomp>
+2024-07-10 08:07:21 | ERROR | stderr |     + [x.to_gradio_chatbot() for x in states]
+2024-07-10 08:07:21 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 130, in to_gradio_chatbot
+2024-07-10 08:07:21 | ERROR | stderr |     return self.conv.to_gradio_chatbot()
+2024-07-10 08:07:21 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/llava/conversation.py", line 206, in to_gradio_chatbot
+2024-07-10 08:07:21 | ERROR | stderr |     msg, image, image_process_mode = msg
+2024-07-10 08:07:21 | ERROR | stderr | ValueError: not enough values to unpack (expected 3, got 2)
+2024-07-10 08:07:22 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:07:22 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:07:22 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:07:22 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:07:22 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:07:22 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:07:22 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:07:22 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:07:22 | ERROR | stderr |     return await future
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:07:22 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:07:22 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:07:22 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:07:22 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:07:22 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 237, in bot_response_multi
+2024-07-10 08:07:22 | ERROR | stderr |     if state0.skip_next:
+2024-07-10 08:07:22 | ERROR | stderr | AttributeError: 'NoneType' object has no attribute 'skip_next'
+2024-07-10 08:19:16 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:19:16 | INFO | stdout | 
+2024-07-10 08:19:16 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:20:02 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:20:02 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:20:02 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:20:02 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1514, in call_function
+2024-07-10 08:20:02 | ERROR | stderr |     prediction = await anyio.to_thread.run_sync(
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:20:02 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:20:02 | ERROR | stderr |     return await future
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:20:02 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:20:02 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 833, in wrapper
+2024-07-10 08:20:02 | ERROR | stderr |     response = f(*args, **kwargs)
+2024-07-10 08:20:02 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_vision_named.py", line 168, in add_text
+2024-07-10 08:20:02 | ERROR | stderr |     states[i] = State(model_selectors[i], is_vision=True)
+2024-07-10 08:20:02 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 102, in __init__
+2024-07-10 08:20:02 | ERROR | stderr |     self.conv = get_conversation_template(model_name)
+2024-07-10 08:20:02 | ERROR | stderr |   File "/home/user/app/src/model/model_adapter.py", line 391, in get_conversation_template
+2024-07-10 08:20:02 | ERROR | stderr |     logger.info("adapter {}", adpater)
+2024-07-10 08:20:02 | ERROR | stderr | NameError: name 'adpater' is not defined. Did you mean: 'adapter'?
+2024-07-10 08:20:03 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:20:03 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:20:03 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:20:03 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:20:03 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:20:03 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:20:03 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:20:03 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:20:03 | ERROR | stderr |     return await future
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:20:03 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:20:03 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:20:03 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:20:03 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:20:03 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 237, in bot_response_multi
+2024-07-10 08:20:03 | ERROR | stderr |     if state0.skip_next:
+2024-07-10 08:20:03 | ERROR | stderr | AttributeError: 'NoneType' object has no attribute 'skip_next'
+2024-07-10 08:20:47 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:20:47 | INFO | stdout | 
+2024-07-10 08:20:47 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:20:57 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:20:57 | INFO | stdout | skip for now
+2024-07-10 08:20:59 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:20:59 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-fire&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f0af01be590>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:20:59 | INFO | gradio_web_server | model_name: llava-fire;model_api_dict: None
+2024-07-10 08:20:59 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:20:59 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-original&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f0af01bed70>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:20:59 | INFO | gradio_web_server | model_name: llava-original;model_api_dict: None
+2024-07-10 08:20:59 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:20:59 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:20:59 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:20:59 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:20:59 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:20:59 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:20:59 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:20:59 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:20:59 | ERROR | stderr |     return await future
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:20:59 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:20:59 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:20:59 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:20:59 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:20:59 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 282, in bot_response_multi
+2024-07-10 08:20:59 | ERROR | stderr |     ret = next(gen[i])
+2024-07-10 08:20:59 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 463, in bot_response
+2024-07-10 08:20:59 | ERROR | stderr |     from src.model.model_llava import inference, inference_by_prompt_and_images
+2024-07-10 08:20:59 | ERROR | stderr | ImportError: cannot import name 'inference_by_prompt_and_images' from 'src.model.model_llava' (/home/user/app/src/model/model_llava.py)
+2024-07-10 08:21:35 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:21:35 | INFO | stdout | 
+2024-07-10 08:21:35 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:21:59 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:21:59 | INFO | stdout | skip for now
+2024-07-10 08:22:00 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:22:00 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-fire&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7feedc05ab60>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:22:00 | INFO | gradio_web_server | model_name: llava-fire;model_api_dict: None
+2024-07-10 08:22:00 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:22:00 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-original&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7feedc05ab00>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:22:00 | INFO | gradio_web_server | model_name: llava-original;model_api_dict: None
+2024-07-10 08:22:00 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:22:00 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:22:00 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:22:00 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:22:00 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:22:00 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:22:00 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:22:00 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:22:00 | ERROR | stderr |     return await future
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:22:00 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:22:00 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:22:00 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:22:00 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:22:00 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 282, in bot_response_multi
+2024-07-10 08:22:00 | ERROR | stderr |     ret = next(gen[i])
+2024-07-10 08:22:00 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 464, in bot_response
+2024-07-10 08:22:00 | ERROR | stderr |     logger.info(f"prompt: {conv.get_prompt()}; images: {images}")
+2024-07-10 08:22:00 | ERROR | stderr |   File "/home/user/app/src/conversation.py", line 169, in get_prompt
+2024-07-10 08:22:00 | ERROR | stderr |     ret += f"{message.strip()}<|eot_id|>"
+2024-07-10 08:22:00 | ERROR | stderr | AttributeError: 'tuple' object has no attribute 'strip'
+2024-07-10 08:24:48 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:24:48 | INFO | stdout | 
+2024-07-10 08:24:48 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:25:03 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:25:03 | INFO | stdout | skip for now
+2024-07-10 08:25:04 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:25:04 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-fire&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7fa73d36d810>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:25:04 | INFO | gradio_web_server | model_name: llava-fire;model_api_dict: None
+2024-07-10 08:25:04 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:25:04 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-original&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7fa73d3962c0>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:25:04 | INFO | gradio_web_server | model_name: llava-original;model_api_dict: None
+2024-07-10 08:25:04 | INFO | gradio_web_server | prompt: <|begin_of_text|><|start_header_id|>system<|end_header_id|>
+
+You are a helpful language and vision assistant. You are able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.<|eot_id|><|start_header_id|>user
+
+<|end_header_id|>
+
+<image>Describe the image<|eot_id|><|start_header_id|>assistant
+
+<|end_header_id|>
+
+; images: ['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']
+2024-07-10 08:25:19 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/spaces/zero/wrappers.py", line 216, in thread_wrapper
+2024-07-10 08:25:19 | ERROR | stderr |     res = future.result()
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 451, in result
+2024-07-10 08:25:19 | ERROR | stderr |     return self.__get_result()
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
+2024-07-10 08:25:19 | ERROR | stderr |     raise self._exception
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/thread.py", line 58, in run
+2024-07-10 08:25:19 | ERROR | stderr |     result = self.fn(*self.args, **self.kwargs)
+2024-07-10 08:25:19 | ERROR | stderr |   File "/home/user/app/src/model/model_llava.py", line 65, in inference_by_prompt_and_images
+2024-07-10 08:25:19 | ERROR | stderr |     image_tensor = process_images(images, image_processor_llava, model_llava.config)
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/llava/mm_utils.py", line 178, in process_images
+2024-07-10 08:25:19 | ERROR | stderr |     image = process_anyres_image(image, image_processor, model_cfg.image_grid_pinpoints)
+2024-07-10 08:25:19 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/llava/mm_utils.py", line 135, in process_anyres_image
+2024-07-10 08:25:19 | ERROR | stderr |     best_resolution = select_best_resolution(image.size, possible_resolutions)
+2024-07-10 08:25:19 | ERROR | stderr | AttributeError: 'str' object has no attribute 'size'
+2024-07-10 08:25:20 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:25:20 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:25:20 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:25:20 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:25:20 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:25:20 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:25:20 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:25:20 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:25:20 | ERROR | stderr |     return await future
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:25:20 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:25:20 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:25:20 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:25:20 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 282, in bot_response_multi
+2024-07-10 08:25:20 | ERROR | stderr |     ret = next(gen[i])
+2024-07-10 08:25:20 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 465, in bot_response
+2024-07-10 08:25:20 | ERROR | stderr |     output_text = inference_by_prompt_and_images(conv.get_prompt(), images)[0]
+2024-07-10 08:25:20 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/spaces/zero/wrappers.py", line 177, in gradio_handler
+2024-07-10 08:25:20 | ERROR | stderr |     raise res.value
+2024-07-10 08:25:20 | ERROR | stderr | AttributeError: 'str' object has no attribute 'size'
+2024-07-10 08:28:47 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:28:47 | INFO | stdout | 
+2024-07-10 08:28:47 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:29:01 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:29:01 | INFO | stdout | skip for now
+2024-07-10 08:29:03 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:29:03 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-fire&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f0b5308b880>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:29:03 | INFO | gradio_web_server | model_name: llava-fire;model_api_dict: None
+2024-07-10 08:29:03 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:29:03 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-original&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f0b656b4be0>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:29:03 | INFO | gradio_web_server | model_name: llava-original;model_api_dict: None
+2024-07-10 08:29:03 | INFO | gradio_web_server | prompt: <|begin_of_text|><|start_header_id|>system<|end_header_id|>
+
+You are a helpful language and vision assistant. You are able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.<|eot_id|><|start_header_id|>user
+
+<|end_header_id|>
+
+<image>describe the image<|eot_id|><|start_header_id|>assistant
+
+<|end_header_id|>
+
+; images: ['/tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png']
+2024-07-10 08:29:10 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/spaces/zero/wrappers.py", line 216, in thread_wrapper
+2024-07-10 08:29:10 | ERROR | stderr |     res = future.result()
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 451, in result
+2024-07-10 08:29:10 | ERROR | stderr |     return self.__get_result()
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
+2024-07-10 08:29:10 | ERROR | stderr |     raise self._exception
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/concurrent/futures/thread.py", line 58, in run
+2024-07-10 08:29:10 | ERROR | stderr |     result = self.fn(*self.args, **self.kwargs)
+2024-07-10 08:29:10 | ERROR | stderr |   File "/home/user/app/src/model/model_llava.py", line 65, in inference_by_prompt_and_images
+2024-07-10 08:29:10 | ERROR | stderr |     image_tensor = process_images(images, image_processor_llava, model_llava.config)
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/llava/mm_utils.py", line 178, in process_images
+2024-07-10 08:29:10 | ERROR | stderr |     image = process_anyres_image(image, image_processor, model_cfg.image_grid_pinpoints)
+2024-07-10 08:29:10 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/llava/mm_utils.py", line 135, in process_anyres_image
+2024-07-10 08:29:10 | ERROR | stderr |     best_resolution = select_best_resolution(image.size, possible_resolutions)
+2024-07-10 08:29:10 | ERROR | stderr | AttributeError: 'str' object has no attribute 'size'
+2024-07-10 08:29:11 | ERROR | stderr | Traceback (most recent call last):
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/queueing.py", line 541, in process_events
+2024-07-10 08:29:11 | ERROR | stderr |     response = await route_utils.call_process_api(
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/route_utils.py", line 276, in call_process_api
+2024-07-10 08:29:11 | ERROR | stderr |     output = await app.get_blocks().process_api(
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1928, in process_api
+2024-07-10 08:29:11 | ERROR | stderr |     result = await self.call_function(
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/blocks.py", line 1526, in call_function
+2024-07-10 08:29:11 | ERROR | stderr |     prediction = await utils.async_iteration(iterator)
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 657, in async_iteration
+2024-07-10 08:29:11 | ERROR | stderr |     return await iterator.__anext__()
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 650, in __anext__
+2024-07-10 08:29:11 | ERROR | stderr |     return await anyio.to_thread.run_sync(
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/to_thread.py", line 56, in run_sync
+2024-07-10 08:29:11 | ERROR | stderr |     return await get_async_backend().run_sync_in_worker_thread(
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 2177, in run_sync_in_worker_thread
+2024-07-10 08:29:11 | ERROR | stderr |     return await future
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/anyio/_backends/_asyncio.py", line 859, in run
+2024-07-10 08:29:11 | ERROR | stderr |     result = context.run(func, *args)
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 633, in run_sync_iterator_async
+2024-07-10 08:29:11 | ERROR | stderr |     return next(iterator)
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/gradio/utils.py", line 816, in gen_wrapper
+2024-07-10 08:29:11 | ERROR | stderr |     response = next(iterator)
+2024-07-10 08:29:11 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_block_arena_named.py", line 282, in bot_response_multi
+2024-07-10 08:29:11 | ERROR | stderr |     ret = next(gen[i])
+2024-07-10 08:29:11 | ERROR | stderr |   File "/home/user/app/src/serve/gradio_web_server.py", line 465, in bot_response
+2024-07-10 08:29:11 | ERROR | stderr |     output_text = inference_by_prompt_and_images(conv.get_prompt(), images)[0]
+2024-07-10 08:29:11 | ERROR | stderr |   File "/usr/local/lib/python3.10/site-packages/spaces/zero/wrappers.py", line 177, in gradio_handler
+2024-07-10 08:29:11 | ERROR | stderr |     raise res.value
+2024-07-10 08:29:11 | ERROR | stderr | AttributeError: 'str' object has no attribute 'size'
+2024-07-10 08:32:58 | INFO | stdout | Running on local URL:  http://0.0.0.0:7860
+2024-07-10 08:32:58 | INFO | stdout | 
+2024-07-10 08:32:58 | INFO | stdout | To create a public link, set `share=True` in `launch()`.
+2024-07-10 08:33:16 | INFO | stdout | moderating image: /tmp/gradio/405613dcd3661394aad4b9b9addbd1743365fabf/screenshot-20240708-164613.png
+2024-07-10 08:33:16 | INFO | stdout | skip for now
+2024-07-10 08:33:17 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:33:17 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-fire&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f081fa88880>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:33:17 | INFO | gradio_web_server | model_name: llava-fire;model_api_dict: None
+2024-07-10 08:33:17 | INFO | gradio_web_server | bot_response. ip: 46.3.240.104
+2024-07-10 08:33:17 | INFO | gradio_web_server | monitor error: HTTPConnectionPool(host='localhost', port=9090): Max retries exceeded with url: /is_limit_reached?model=llava-original&user_id=46.3.240.104 (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f081fa89db0>: Failed to establish a new connection: [Errno 111] Connection refused'))
+2024-07-10 08:33:17 | INFO | gradio_web_server | model_name: llava-original;model_api_dict: None
+2024-07-10 08:33:17 | INFO | gradio_web_server | prompt: <|begin_of_text|><|start_header_id|>system<|end_header_id|>
+
+You are a helpful language and vision assistant. You are able to understand the visual content that the user provides, and assist the user with a variety of tasks using natural language.<|eot_id|><|start_header_id|>user
+
+<|end_header_id|>
+
+<image>Describe the image<|eot_id|><|start_header_id|>assistant
+
+<|end_header_id|>
+
+; images: ['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']
+2024-07-10 08:33:25 | ERROR | stderr | /usr/local/lib/python3.10/site-packages/transformers/generation/configuration_utils.py:392: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `temperature`.
+2024-07-10 08:33:25 | ERROR | stderr |   warnings.warn(
+2024-07-10 08:33:25 | ERROR | stderr | /usr/local/lib/python3.10/site-packages/transformers/generation/configuration_utils.py:397: UserWarning: `do_sample` is set to `False`. However, `top_p` is set to `0.9` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `top_p`.
+2024-07-10 08:33:25 | ERROR | stderr |   warnings.warn(
+2024-07-10 08:33:31 | INFO | gradio_web_server | hello
+2024-07-10 08:33:31 | INFO | gradio_web_server | The image captures a moment of proposal between two individuals. The man, dressed in a classic gray suit, is standing upright with his arms outstretched, holding a ring in his left hand. His posture and facial expression suggest he is in the act of proposing. The woman, wearing a vibrant red dress, is kneeling on the ground with her hands clasped together, indicating her readiness to accept the proposal. The simplicity of the background focuses the viewer's attention on the interaction between the two characters. The image is devoid of any text or additional context, allowing the viewer to focus solely on the emotional exchange between the man and the woman.