Firebase Studio constantly giving “hits a snag” on every prompt

Gemini API
, , ,
1

I wanted to make some modifications to my Firebase Studio project, but every time I submit a prompt, it returns a constant ‘Sorry, I hit a snag.’ error. This happens regardless of the prompt content (even on standard code analysis or minimal modification like font colour change or debugging queries).

image
image470×177 7.82 KB

here is the exact error:
Sorry, I hit a snag. Please try again shortly or modify your prompt.

[GoogleGenerativeAI Error]: Error fetch

ing from https://monospace-pa.googleapis.com/v1/models/gemini-3-flash-preview:streamGenerateContent?alt=sse: [400 Bad Request] Unable to submit request because Thinking_config.include_thoughts is only enabled when thinking is enabled.. Learn more: Generate content with the Gemini API in Vertex AI  |  Generative AI on Vertex AI  |  Google Cloud Documentation

Context & What I’ve Tried:

  • This happens on absolutely any prompt I enter (even simple text modifications), so it doesn’t seem to be a content or safety block.

  • I’ve tried giving prompt after 24+ hours , but still the same error

  • I have tried refreshing the page, but the error persists.

  • There is no visible option in my UI to toggle “Thinking” or “Reasoning” configurations to fix this configuration mismatch.

How do I solve it??

2

Has anyone else run into this issue in Firebase Studio?

[GoogleGenerativeAI Error]: Error fetching from https://monospace-pa.googleapis.com/v1/models/gemini-3-flash-preview:streamGenerateContent?alt=sse: [400 Bad Request] Unable to submit request because Thinking_config.include_thoughts is only enabled when thinking is enabled.. Learn more: Generate content with the Gemini API in Vertex AI  |  Generative AI on Vertex AI  |  Google Cloud Documentation

A few observations from my testing:

  • Started appearing suddenly without code changes.

  • Occurs across multiple Firebase Studio workspaces.

  • Searching the codebase shows no explicit includeThoughts configuration.

  • Refreshing/reopening workspaces did not resolve it.

  • The error appears to originate from the Firebase Studio AI Assistant request layer.

I contacted Firebase Support and received the following response:

“We were able to reproduce the issue and are actively working to resolve it.”

So it appears to be a known issue that Firebase has acknowledged.

I’m posting this in case others encounter the same error. Has anyone found a temporary workaround while waiting for the official fix?

3

I have this problem too, maybe they will fix it, check the status here https://status.firebase.google.com/

4

Hello, I’ve had the same problem since Saturday night. It’s happened to me before; last time I managed to fix it by changing the IDX model. This time it’s different. I also tried creating an .aiexclude file, but nothing worked… I’m still looking for a solution.

5

I’m having the exact same problem as @M888; it stopped working on Saturday. I had attributed it to a token overrun, but that wasn’t the case. We hope for a timely resolution. We’re still waiting.

6

I’m having the same issue, has anyone found a solution yet?

7

Have you managed to find a solution yet?

8

Until this issue is resolved, you can temporarily use Gemini support as follows:

  1. Switch to ‘Code mode’.

  2. Click the [+] icon at the top and create a “New chat”.

  3. Set the chat type to “Agent (Auto run)”.

Now you can provide the relevant instructions.

10

Same problem here… all day trying to fix it…

11

It is working fine now. I believe they have resolved the issue.

12

The issue appears to be resolved and the system is working normally again without needing any workarounds. Thanks everyone!