Frequent 503 "The model is overloaded" errors on Gemini 2.5 Flash

Hi everyone,

I’m running Gemini 2.5 Flash in production (for multiple apps) and I keep hitting this error:

503 - The model is overloaded. Please try again later.

This makes my apps unusable for end users, since requests randomly fail. From what I’ve seen, this issue is not just on my side, many other developers are running into the same problem. AND ITS NOT ABOUT RATE LIMIT. The probleme appeared approximately one month ago.

It’s extremely disruptive for production workloads, and it would be great if this could be prioritized and fixed as soon as possible.

Is there any official update or timeline on when this will be resolved?

Thanks!

7 Likes

I am also getting this error frequently.

1 Like

Hey everyone,

Any plans on looking on this error? More than one week having the same error with Gemini 2.5 pro (of course, not the free tier). Already reported that on github and no responses °_°

1 Like

No repsonse at all. I tried this forum, Google dev discord, tagging Google devs on X, nobody cares. And its geting worse since they are pushing new models using more compute

2 Likes

This is such a shame-no outage notification, just ignored errors, and when everyone is facing this issue, there’s no response from any side.

2 Likes

I’m getting this quite a bit on gemini-2.5-pro :confused:

1 Like

Apparently this is “the best time to use Gemini”. Now its clear they don’t care at all:

3 Likes

Hi We are facing issue google.genai.errors.ServerError: 503 UNAVAILABLE. {‘error’: {‘code’: 503, ‘message’: ‘The model is overloaded. Please try again later.’, ‘status’: ‘UNAVAILABLE’}}. Kindly advise on further steps

1 Like

Same errors, now even on paid API. What is happening for nearly 1 month with Gemini APIs?

Hi Team, we are facing the issue again. kindly let us know what we need to do

We also have issue with the Gemini 2.5 Flash model, which has been returning frequent 503 Service Unavailable errors since October 10th

i am also facing the same issue even for the small requests it does this and if it is half complete it consumes tokens and the remaining half failed it is way to frustrating i tried retry various times but it is resulting the same issue please do check

Same situation as you. Just getting this error message randomly and frequently:

Service unavailable - try again later or consider setting this node to retry automatically (in the node settings)

[GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent: [503 Service Unavailable] The model is overloaded. Please try again later.

I’m still getting this issue

Nothing has changed and they keep blind about this. Nice.

Using 2.5 flash and am using it really less due to being busy with other stuff, how constant is it?

If it is a sporadic issue, does it last more than a couple of hours?

Is there a workaround? Like using another ai model( openai, claude) as a backup service?

Or we are just as helpless as a farmer waiting for rains?

I’m facing the same issue with AI Studio trying to use 2.5 Flash.
It’s quite sad really, because my tests were great and now, when I want to use it with real data, it just 503th itself…
Should I just wait?