# How to fix this issue **URL:** \ **Category:** AI\ **Tags:** troubleshooting\ **Created:** [May 23, 2025, 2:49pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278 "2025-05-23T14:49:11Z")\ **Posts on this page:** 8\ **Page:** 1
**Author:** ![Color\_Hut\_Official](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/color_hut_official/32/6818_2.png) [@Color\_Hut\_Official](https://community.firebasestudio.dev/u/Color_Hut_Official)\ **Post date:** [May 23, 2025, 2:49pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/1 "2025-05-23T14:49:11Z")
![WhatsApp Image 2025-05-21 at 15.22.24_0771891a](https://us1.discourse-cdn.com/flex002/uploads/idxcommunity/original/2X/9/9fe3bb78d57c1effffe4550123b26ed309d2c126.jpeg) ---
**Author:** ![DaC](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/dac/32/6866_2.png) [@DaC](https://community.firebasestudio.dev/u/DaC)\ **Post date:** [May 23, 2025, 3:03pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/2 "2025-05-23T15:03:16Z")
Reset the project VM, wait a few minutes and start again. It won’t erase your data. ![image](https://us1.discourse-cdn.com/flex002/uploads/idxcommunity/original/2X/d/da503a90ba48dd1dac5bcad4b290bb993c705a2e.png) ---
**Author:** ![t\_rec](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/t_rec/32/6915_2.png) [@t\_rec](https://community.firebasestudio.dev/u/t_rec)\ **Post date:** [May 23, 2025, 3:19pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/3 "2025-05-23T15:19:28Z")
that doesn’t solve it anymore, at least not for me. this is a new “screen of death”, it comes back the second the Prototyper begins even remotely talking about code in its response. it’s extremely frustrating. ---
**Author:** ![DaC](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/dac/32/6866_2.png) [@DaC](https://community.firebasestudio.dev/u/DaC)\ **Post date:** [May 23, 2025, 3:21pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/4 "2025-05-23T15:21:59Z")
their severs must be overloaded ---
**Author:** ![t\_rec](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/t_rec/32/6915_2.png) [@t\_rec](https://community.firebasestudio.dev/u/t_rec)\ **Post date:** [May 23, 2025, 3:27pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/5 "2025-05-23T15:27:50Z")
that’s really none of our concern! why are we allowed in? why give us false hope? and how are we to know if they are back up to regular speed or whatever, if they don’t show that it’s been DOWN for what feels like ages. how are we supposed to know what is going on and when and what is going to be done? ---
**Author:** ![DaC](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/dac/32/6866_2.png) [@DaC](https://community.firebasestudio.dev/u/DaC)\ **Post date:** [May 23, 2025, 3:36pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/6 "2025-05-23T15:36:18Z")
since this is currently working 100% on my side (it wasn’t a few hours back), I can only think this is related to VM location, somehow I managed to be put in a VM location that is working normally… but I suspect at some point they will solve it, as it doesn’t makes sense for Google either to give such experience for the users as their service is picking up… I would say they’re the party most interested in getting us working in their platform… they’ll probably allocate more resources to it now that the usage seems to be increasing… I was actually perplexed how firebase studio wasn’t being raved on the internet yet… I guess it will very shortly… when some youtber makes a sweet review of it, then we’ll have even more problems using it, or they’ll start charging/limiting it. ---
**Author:** ![t\_rec](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/t_rec/32/6915_2.png) [@t\_rec](https://community.firebasestudio.dev/u/t_rec)\ **Post date:** [May 23, 2025, 3:40pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/7 "2025-05-23T15:40:26Z")
everyody, including the Firebase/Studio staff, should be paying attention to what [https://app.emergent.sh/](https://app.emergent.sh/) is doing: allowing us to manually go back up to a previous build and _delete_ either the code and chat history OR **only** the chat history, therefore refreshing and lightening the load of the instance and starting fresh with a simple contextual prompt of the entire instance before cleaning the slate, provided within the instance itself, and we can go on forever. I ran a project on such strings of reinitialized instances for 3 days straight during a long weekend, and I shit you not, it did not crash ONCE. not ONCE. ---
**Author:** ![DaC](https://sea2.discourse-cdn.com/flex002/user_avatar/community.firebasestudio.dev/dac/32/6866_2.png) [@DaC](https://community.firebasestudio.dev/u/DaC)\ **Post date:** [May 23, 2025, 3:49pm UTC](https://community.firebasestudio.dev/t/how-to-fix-this-issue/13278/8 "2025-05-23T15:49:29Z")
yeah, this would certainly be good to have, although we can kind do it the backend, but indeed something they should consider implement on the graphic interface. I just hope they don’t restrict context window size, or else will become another cursor like, which I find pretty bad with context, which is one of the main strength from Gemini that I really enjoy. > [@Temporary workaround for error due to exceeding input token count (unofficial)](https://community.firebasestudio.dev/t/temporary-workaround-for-error-due-to-exceeding-input-token-count-unofficial/12457): > > Good news for those who are having trouble with the input token count exceeding the limit! I’ve found a temporary solution. However, please note that this is not an official solution. The error discussed in this article [GoogleGenerativeAI Error]: Error fetching from https://monospace-pa.googleapis.com/v1/models/gemini-2.5-pro-preview-03-25:streamGenerateContent?alt=sse: [400 Bad Request] The input token count (1053137) exceeds the maximum number of tokens allowed (1048576).