Gemini API 404 Not Found for gemini-pro-vision on v1beta endpoint, despite model being in list_models

Gemini API
,
1

Hello,

I’m encountering a persistent 404 Not Found error when trying to access standard Gemini models. My requests are consistently being routed to a v1beta endpoint where models like gemini-pro-vision and gemini-pro are not found, even though my list_models() call shows they are available.

The key contradiction is:

  • A script to list available models successfully shows models/gemini-pro-vision as an option that supports generateContent.

  • A script to use that exact model (generate_content) fails with a 404 Not Found error.

Troubleshooting Steps I’ve Already Taken: I have already tried the following steps without any change in the error:

  • Forcefully reinstalling the google-generativeai library to ensure it’s the latest version.

  • Running the scripts in a completely clean, new Python virtual environment (venv).

  • Generating a brand new API key in a completely new Google AI Studio project.

  • Trying different model names, including gemini-pro, gemini-1.5-flash-latest, and gemini-pro-vision.

  • Using the full model path models/gemini-pro-vision.

Any help in understanding why my project is stuck on the v1beta endpoint or why generate_content cannot find a model that list_models can see would be greatly appreciated. Thank you.

3

Hello,

welcome to the Forum,

Could you please share relevant code so that we can reproduce your issue and come up with an accurate response?

4

Hi Lalit,

I have attached a python script. To give you some context I am trying to automate the process by getting Gemini to recognise images and create a CSV that contains the relevant identification.


import google.generativeai as genaifrom PIL import Imageimport os

API_KEY = “YOUR_API_KEY_HERE”

IMAGE_FOLDER = r"C:\Path\To\Your\Images"IMAGE_FILENAME = “example_image.jpg”

try:genai.configure(api_key=API_KEY)image_path = os.path.join(IMAGE_FOLDER, IMAGE_FILENAME)

print(f"Attempting to process one image: {image_path}")

if not os.path.exists(image_path):
    print(f"\n--- ERROR! ---")
    print(f"The image file was not found.")
    exit()

# Using the full model name that is failing
model = genai.GenerativeModel('models/gemini-pro-vision')

img = Image.open(image_path)
prompt = "Describe this image in one sentence."

response = model.generate_content([prompt, img])

print("\n--- SUCCESS! ---")
print(response.text)

except Exception as e:print(“\n— ERROR! —”)print(f"The API call failed with the following error:\n{e}")

It then produces the following error:

--- ERROR! ---
The API call failed with the following error: 404 models/gemini-pro-vision is not found for API version v1beta, or is not supported for generateContent. Call ListModels to see the list of available models and their supported methods.

Thanks.

6

Hello,

If you run the code provided, you will notice that there is no model named models/gemini-pro-vision. Could you recheck and retry with correct model name?

import google.genai as genai

client = genai.Client(api_key="GEMINI_API_KEY")
models = client.models.list()
for model in models:
    print(model.name)