# Attribute "memory" triggers an api error?

**URL:** <https://community.deeplearning.ai/t/attribute-memory-triggers-an-api-error/663218>\
**Category:** Multi AI Agent Systems with crewAI\
**Created:** [July 15, 2024, 3:29am UTC](https://community.deeplearning.ai/t/attribute-memory-triggers-an-api-error/663218 "2024-07-15T03:29:16Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![fairydora](https://avatars.discourse-cdn.com/v4/letter/f/e99b99/32.png) [@fairydora](https://community.deeplearning.ai/u/fairydora)\
**Post date:** [July 15, 2024, 3:29am UTC](https://community.deeplearning.ai/t/attribute-memory-triggers-an-api-error/663218/1 "2024-07-15T03:29:16Z")

</div>

Hi,  
I’ve got a paid account for chatGPT but not for its API which requires a separate subscription.  
To run the exercises locally (L2 and L3), I changed the default model (chatGPT) to openhermes from Ollama.

from langchain.llms import Ollama  
llm = Ollama(model=“openhermes”)

In each agent, I’ve added the additional attribute llm=llm to override the default chatGPT.  
Exercise L2 worked straight away, but L3 displayed this error:

_openai.RateLimitError: Error code: 429 - {‘error’: {‘message’: ‘You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: [https://platform.openai.com/docs/guides/error-codes/api-errors.](https://platform.openai.com/docs/guides/error-codes/api-errors.)’, ‘type’: ‘insufficient\_quota’, ‘param’: None, ‘code’: ‘insufficient\_quota’}}_

After debugging I determined that the attribute “memory” in crew is responsible - when I comment it out, L3 works with openhermes, but when I add it back, I get the error back, although I’m not using chatGPT.  
Would you be able to give me a bit of help with this ?..