Bug描述

from llama_index.llms.nvidia import NVIDIA 出现以下错误

版本

llama_index (0.10.55) llama_index_llms_openai (0.1.25) ,llama-index-llms-nvidia (0.1.4)

重现步骤

pip install llama-index llama-index-llms-nvidia llama-index-llms-openai

相关日志/回溯

The full log is as follows.

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
Cell In[88], line 1
----> 1 from llama_index.llms.nvidia import NVIDIA

File /opt/conda/lib/python3.11/site-packages/llama_index/llms/nvidia/__init__.py:1
----> 1 from llama_index.llms.nvidia.base import NVIDIA
      3 __all__ = ["NVIDIA"]

File /opt/conda/lib/python3.11/site-packages/llama_index/llms/nvidia/base.py:16
     10 from llama_index.core.bridge.pydantic import PrivateAttr, BaseModel
     11 from llama_index.core.base.llms.generic_utils import (
     12     get_from_param_or_env,
     13 )
---> 16 from llama_index.llms.openai_like import OpenAILike
     18 DEFAULT_MODEL = "meta/llama3-8b-instruct"
     19 BASE_URL = "https://integrate.api.nvidia.com/v1/"

File /opt/conda/lib/python3.11/site-packages/llama_index/llms/openai_like/__init__.py:1
----> 1 from llama_index.llms.openai_like.base import OpenAILike
      3 __all__ = ["OpenAILike"]

File /opt/conda/lib/python3.11/site-packages/llama_index/llms/openai_like/base.py:20
     14 from llama_index.core.constants import DEFAULT_CONTEXT_WINDOW
     15 from llama_index.core.base.llms.generic_utils import (
     16     async_stream_completion_response_to_chat_response,
     17     completion_response_to_chat_response,
     18     stream_completion_response_to_chat_response,
     19 )
---> 20 from llama_index.llms.openai.base import OpenAI, Tokenizer
     21 from transformers import AutoTokenizer
     24 class OpenAILike(OpenAI):

ModuleNotFoundError: No module named 'llama_index.llms.openai.base'; 'llama_index.llms.openai' is not a package