Issue with count_tokens in langchain_aws Using Claude on Bedrock

[MehyarWealthy]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

def summarize_with_claude(text_chunks):
    model = ChatBedrock(
        client=bedrock_runtime,
        model_id=inference_profile_arn,  
    )

    # Define a custom prompt for summarization
    prompt_template = ChatPromptTemplate.from_messages([
        ("system", "You are a helpful summarizer of medical scientific papers."),
        ("human", "Write a concise summary of the following text:\n{text}\nCONCISE SUMMARY:")
    ])

    # Convert chunks into LangChain Document objects
    docs = [Document(page_content=chunk) for chunk in text_chunks]

    # Load the summarization chain
    chain = load_summarize_chain(model, chain_type="map_reduce", return_intermediate_steps=False)

    # Run the summarization chain
    summary = chain.invoke(docs)
    return summary


  File "C:\Users\14384\Desktop\..\Projects\med-summarizer\summerizer.py", line 105, in <module>
    summary = summarize_with_claude([test_text])
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\14384\Desktop\Wealthy-Technlogy\Projects\med-summarizer\summerizer.py", line 65, in summarize_with_claude
    summary = chain.invoke(docs)
              ^^^^^^^^^^^^^^^^^^
...
  File "C:\Users\14384\Desktop\..\Projects\med-summarizer\summerizer\Lib\site-packages\langchain_aws\utils.py", line 25, in get_num_tokens_anthropic
    return client.count_tokens(text=text)

Description

I'm encountering an issue when trying to use langchain_aws with Claude on AWS Bedrock. The error occurs when LangChain attempts to count tokens using client.count_tokens(text=text).

System Info

Python 3.11.9
boto3==1.36.15
botocore==1.36.15
langchain==0.3.17
langchain-aws==0.2.12
langchain-core==0.3.34
langchain-text-splitters==0.3.6
langsmith==0.3.6