ValueError: could not determine the shape of object type 'VideoMetadata'

[markmochi200]

Reminder

• I have read the above rules and searched the existing issues.

System Info

• llamafactory version: 0.9.5.dev0
• Platform: Linux-6.8.0-94-generic-x86_64-with-glibc2.35
• Python version: 3.12.12
• PyTorch version: 2.10.0+cu128 (GPU)
• Transformers version: 5.3.0.dev0
• Datasets version: 4.0.0
• Accelerate version: 1.11.0
• PEFT version: 0.18.1
• GPU type: NVIDIA GeForce RTX 2080 Ti
• GPU number: 1
• GPU memory: 21.48GB
• TRL version: 0.24.0
• vLLM version: 0.16.1rc1.dev198+g70c73df69
• Default data directory: not detected

Reproduction

[WARNING|2026-03-04 14:06:48] llamafactory.chat.hf_engine:155 >> There is no current event loop, creating a new one.
Traceback (most recent call last):
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/queueing.py", line 759, in process_events
    response = await route_utils.call_process_api(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/route_utils.py", line 354, in call_process_api
    output = await app.get_blocks().process_api(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/blocks.py", line 2191, in process_api
    result = await self.call_function(
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/blocks.py", line 1710, in call_function
    prediction = await utils.async_iteration(iterator)
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/utils.py", line 760, in async_iteration
    return await anext(iterator)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/utils.py", line 751, in __anext__
    return await anyio.to_thread.run_sync(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/anyio/to_thread.py", line 63, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/anyio/_backends/_asyncio.py", line 2502, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/anyio/_backends/_asyncio.py", line 986, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/utils.py", line 734, in run_sync_iterator_async
    return next(iterator)
           ^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/gradio/utils.py", line 898, in gen_wrapper
    response = next(iterator)
               ^^^^^^^^^^^^^^
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/webui/chatter.py", line 218, in stream
    for new_text in self.stream_chat(
                    ^^^^^^^^^^^^^^^^^
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/chat/chat_model.py", line 135, in stream_chat
    yield task.result()
          ^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/concurrent/futures/_base.py", line 456, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/chat/chat_model.py", line 150, in astream_chat
    async for new_token in self.engine.stream_chat(
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/chat/hf_engine.py", line 394, in stream_chat
    stream = self._stream_chat(*input_args)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/anaconda3/envs/GrokkingTrf/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 124, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/chat/hf_engine.py", line 281, in _stream_chat
    gen_kwargs, _ = HuggingfaceEngine._process_args(
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/lx3005/Desktop/GrokkingTrf/LlamaFactory/src/llamafactory/chat/hf_engine.py", line 190, in _process_args
    value = torch.tensor(value)
            ^^^^^^^^^^^^^^^^^^^
ValueError: could not determine the shape of object type 'VideoMetadata'

Others

I am chatting with Qwen3.5-0.8B through webui, and run into issues when inferring with video