run_flow.py 执行步骤过多时会报错token长度超限

[ZhXZhao]

config
[llm]
model = "qwen-max" # The LLM model to use
base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1" # API endpoint URL
max_tokens = 8192 # Maximum number of tokens in the response
temperature = 0.0

[llm.vision]
model = "qwen-vl-max" # The vision model to use
base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1" # API endpoint URL for vision model
max_tokens = 8192 # Maximum number of tokens in the response
temperature = 0.0 # Controls randomness for vision model

当执行步骤大于20时，会一直报错
OpenAI API error: Error code: 400 - {'error': {'code': 'invalid_parameter_error', 'param': None, 'message': '<400> InternalError.Algo.InvalidParameter: Input tokens length is 29618. Output tokens length is 8192. Range of total length should be less than or equal to 32768', 'type': 'invalid_request_error'}

这个是因为执行步骤过多，导致上下文长度超限了吗？
run_flow.py应是多agent架构，每个agent都有32768的输入长度token，这个出现token超限的原因是什么？

[HaruHoo]

我也遇到这个问题了，不知和Memory有没有关系:)