Copilot Agent Error Code

(2) Share

Report

Posted on by CS-19030835-0

I'm suddenly having trouble with my agent presenting this error code:

Error Message: The request is resulting in too much data to handle, please evaluate the amount of data being returned by your actions.[Learn more]https://learn.microsoft.com/en-us/microsoft-copilot-studio/error-codes?tabs=webApp#toomuchdatatohandleError Code: OpenAIModelTokenLimit

A week ago I did not have this issue. We have not changed the amount of data in our knowledge sources.

I can see when the Agent is searching the available sources, the same (relevant) document is listed many many times, as well as other documents.

What might be the cause of this, and is there an easy fix?

Categories:

Building Copilot Studio chatbots in Microsoft Teams

I have the same question (0)

All responses (3)

Answers (0)

Sort by

Suggested answer

Valantis 6,163 on at

Like (1)

Report
Copy link

Link copied!

Hi @CS-19030835-0,
The Microsoft error codes doc confirms this: the OpenAIModelTokenLimit error happens when the total request size exceeds the model's limit.

The request includes user input, conversation history, and everything retrieved from knowledge sources and tools.

The symptom you are seeing the same document appearing many times in the search results suggests the agent is pulling too many chunks from the same document during retrieval, bloating the context before it even reaches the model.

Things to check and try:

1. Scope down your knowledge sources. If you are pointing at a full SharePoint site, switch to a specific document library or folder instead. Narrowing the scope reduces how many chunks get retrieved.
2. Check if any of your documents have grown in size recently. Even if the number of files has not changed, a larger file means more chunks being returned.
3. If you have any tools or actions that return large outputs, scope them down to only the necessary fields. The error doc specifically calls this out.
4. Start a fresh conversation when testing. Long conversation history also adds to the token count and can push you over the limit.

Ref: https://learn.microsoft.com/en-us/microsoft-copilot-studio/error-codes

Best regards,

Valantis

✅ If this helped solve your issue, please Accept as Solution so others can find it quickly.

❤️ If it didn’t fully solve it but was still useful, please click “Yes” on “Was this reply helpful?” or leave a Like :).

🏷️ For follow-ups @Valantis.

📝 https://valantisond365.com/

Was this reply helpful? Yes No
Suggested answer

Sayali-MSFT Microsoft Employee on at

Like (1)

Report
Copy link

Link copied!

Hello CS-19030835-0 ,

The OpenAIModelTokenLimit error you're encountering is most likely not due to any increase in your knowledge source data, but rather a change in how Copilot Studio’s retrieval system is ranking and injecting grounding content into the LLM prompt. Recent backend updates have caused the agent to retrieve multiple semantically similar chunks from the same document—sometimes dozens—which silently inflates the token payload sent to the model and exceeds its processing limit. This is why you're now seeing the same document listed repeatedly during knowledge search. To mitigate this, you can limit the number of knowledge sources used per topic, split large documents like job description master files into smaller role-specific files, remove repeated headers or footers from uploaded content, or—preferably—pass only the relevant job description into the agent via Power Automate and disable unbounded knowledge search for scoring altogether. This significantly reduces the size of grounding data and improves reliability for token-sensitive tasks like candidate evaluation.

Was this reply helpful? Yes No
Valantis 6,163 on at

Like (1)

Report
Copy link

Link copied!

Hi @CS-19030835-0,

Just wanted to check in and see if everything is working now. If you still need any help, feel free to let me know.

Also, if the issue is resolved, it would be great if you could mark the answer as solved so others with the same question can find it easily.

Thanks and have a great day!

Was this reply helpful? Yes No