I have tried omitting the “video part” of the content array so that I’m only sending the prompt, and this works. I have also tried using an obviously incorrect file uri, and this results in a different error (invalid uri).
Curious if anybody else is experiencing the same issue. Seems like it may be an issue with Google’s API if I haven’t missed anything here.
Update: I tried switching the model and so far I don’t get this error with gemini-1.5-pro and gemini-1.5-flash-8b. The issue seems to be specifically with 1.5-flash.
Using, this documentation, I had no issue with gemini-1.5-pro but encountered your error with `gemini-1.5-flash and different versions of it. Let me further into this and get back to you.
I’m getting the same RESOURCE_PROJECT_INVALID error with gemini-1.5-flash:streamGenerateContent
generateContent is working fine, as is gemini-2.0-flash-exp
Hello everyone. This is issue has been reported to the product team and I will provide updates once they respond. Thank you for your patience. @Logan_Kilpatrick You can reproduce the issue by following this Prompt with a video and text example, using gemini-1.5-flash model.
curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:streamGenerateContent?key=[API KEY]" \
-H 'Content-Type: application/json' \
-d '{ "system_instruction": {"parts":{ "text": "You are a cat. Your name is Neko."}},"contents": {"parts": {"text": "Hello there"}}}'
This gives the following error:
[{
"error": {
"code": 400,
"message": "Developer instruction is not enabled for models/gemini-1.5-flash",
"status": "INVALID_ARGUMENT"
}
}
]
It started happening yesterday (1/8/2025) at around 10am pacific. It doesn’t seem to affect everyone’s account, so you might have to cycle through some different API keys to see it.
Hey folks, we’ve pushed a fix for both the video uploads via the File API and the streamGenerateContent issues, and they should be live soon. Please let us know if you continue to experience these problems