Whenever the response uses a bit more tokens, the issue shown in the image occurs. I did some testing and found that this seems to happen when answering coding-related questions at around 1,000 lines of code. I’m quite certain that this limitation did not exist before. I hope this limit can be increased or relaxed.
Whenever the response gets a little longer and uses more tokens, the issue shown in the image occurs
Hello @PlaidMrdeer,
Could you please try again and, if it still happens, share:
- The prompt you used (or a shorter version that still triggers it, with nothing sensitive)
- Any files or public resources you attached
- Roughly when it started and how often it happens
@Payal_Sharma2 It seems that the issue has nothing to do with the prompts or uploaded files. No matter how I change the uploaded files or the prompts, the problem still occurs whenever the response contains a large amount of code. The frequency of occurrence is inconsistent, so I’m not sure whether it could be a network issue. However, this problem had never occurred before.
Hello @PlaidMrdeer,
Even if this occurs across different prompts and files, we need at least one concrete example to reproduce and look into it. Could you please share:
- One specific prompt (or a minimal example prompt) that triggered the
Recitation blockmessage - The exact model selected in Google AI Studio
- A HAR file (capture steps) recorded when the
Recitation blockmessage appears