[Python] Google Generative AI: 404 "models/gemini-1.5-flash is not found" error when calling...

I have a Flask application that integrates with Google Generative AI (Gemini) and Google Calendar. It used to work fine, but recently I started running into two issues:

1. Model not found (404) error

Whenever my code calls the Gemini API, I get:

ERROR:gemini_parser:Error extracting candidate info: 404 models/gemini-1.5-flash 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.
ERROR:gemini_parser:Error scoring candidate: 404 models/gemini-1.5-flash is not found for API version v1beta...


This happens for every resume processed, and all candidates get marked as "Parse Error" with a default score.

Here’s how I initialize the model:

import google.generativeai as genai

genai.configure(api_key="MY_API_KEY")
model = genai.GenerativeModel("gemini-1.5-flash")

response = model.generate_content("Test prompt")
print(response.text)


This exact code worked before, but now I get the error above.

1. SSL errors in logs

At startup, I also see warnings like:

WARNING:tools:SSL/Connection error on attempt 1: [SSL: WRONG_VERSION_NUMBER] wrong version number (_ssl.c:2580)
WARNING:tools:SSL/Connection error on attempt 2: [SSL: DECRYPTION_FAILED_OR_BAD_RECORD_MAC] decryption failed or bad record mac (_ssl.c:2580)
ERROR:tools:Error checking availability: The read operation timed out


Despite these, my Flask app continues to run on http://127.0.0.1:5000 and processes requests.


What I tried:

• 
  Restarted the app and re-authenticated with Google API.
  
  
• 
  Verified my API key and credentials are set.
  
  
• 
  Tested genai.list_models() — it returns a generator, but I’m not sure if gemini-1.5-flash is still in the list.
  
  
• 
  Tried different networks to rule out connectivity.
  

My Questions:

1. 
   Why is models/gemini-1.5-flash returning 404 now?
   
   Did Google deprecate or rename it?
   
   Is it only available under certain API versions (v1 vs v1beta)?
   
   
2. 
   Could API quota exhaustion cause this behavior?
   
   
3. 
   What’s the correct way to check the currently available Gemini models?
   
   
4. 
   Are the SSL errors related to the 404 issue, or just a separate local misconfiguration (Flask dev server using HTTP instead of HTTPS)?
   

Any guidance would be appreciated

Continue reading...