deepseek-coder-33b-instruct model with openai got "InvalidChunkLength" error

Use FastChat to start the deepseek-coder-33b-instruct model, send a stream request and got an error response.
If set stream=False, you can print a good response
If change to other models, it also works with stream
Start cmd:

python3 -m fastchat.serve.controller
python3 -m fastchat.serve.openai_api_server --host 0.0.0.0 --port 8000
python3 -m fastchat.serve.model_worker --model-path /home/ubuntu/models/deepseek-coder-33b-instruct --num-gpus=2 --gpus=0,1 --max-gpu-memory=46GB --model-names=deepseek-coder-33b-instruct

Below python code for test:

import openai

if __name__ == "__main__":
    openai.api_base = "http://model_addr:8000/v1"
    openai.api_key = "none"

    response = openai.ChatCompletion.create(
        model="deepseek-coder-33b-instruct",
        temperature=0.6,
        top_p=1,
        messages=[{"role": "user", "content": "Hello"}],
        stream=True
    )
    print(response)
    for chunk in response:
        print(chunk, end="", flush=True)

error log:

<generator object EngineAPIResource.create.<locals>.<genexpr> at 0x1030ae900>
Traceback (most recent call last):
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 712, in _error_catcher
    yield
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 1071, in read_chunked
    self._update_chunk_length()
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 1006, in _update_chunk_length
    raise InvalidChunkLength(self, line) from None
urllib3.exceptions.InvalidChunkLength: InvalidChunkLength(got length b'', 0 bytes read)

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/requests/models.py", line 816, in generate
    yield from self.raw.stream(chunk_size, decode_content=True)
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 931, in stream
    yield from self.read_chunked(amt, decode_content=decode_content)
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 1100, in read_chunked
    self._original_response.close()
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/contextlib.py", line 137, in __exit__
    self.gen.throw(typ, value, traceback)
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/urllib3/response.py", line 729, in _error_catcher
    raise ProtocolError(f"Connection broken: {e!r}", e) from e
urllib3.exceptions.ProtocolError: ("Connection broken: InvalidChunkLength(got length b'', 0 bytes read)", InvalidChunkLength(got length b'', 0 bytes read))

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/Users/yangglei/work/projects/go/src/python/test-continue-2.py", line 34, in <module>
    for chunk in response:
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/openai/api_resources/abstract/engine_api_resource.py", line 166, in <genexpr>
    return (
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/openai/api_requestor.py", line 692, in <genexpr>
    return (
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/openai/api_requestor.py", line 115, in parse_stream
    for line in rbody:
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/requests/models.py", line 865, in iter_lines
    for chunk in self.iter_content(
  File "/Users/yangglei/miniconda3/envs/my-env/lib/python3.9/site-packages/requests/models.py", line 818, in generate
    raise ChunkedEncodingError(e)
requests.exceptions.ChunkedEncodingError: ("Connection broken: InvalidChunkLength(got length b'', 0 bytes read)", InvalidChunkLength(got length b'', 0 bytes read))

Start by running the listed controller, OpenAI API server, and model worker commands with deepseek-coder-33b-instruct, then reproduce the streaming request from the Python example. Trace the FastChat OpenAI API streaming path and compare it with a non-streaming request and another model. Done means the affected model streams successfully without the reported chunked-encoding error.