Configuring openai is not working

[lsdi2016]

Hello All

In my .env, i enabled openai, but when i execute, it will always point to GeminiAI. Please help us understand, how i can change the engine.

PaperBanana - Generating Methodology Diagram │
│ │
│ VLM: gemini / gemini-2.0-flash │
│ Image: google_imagen / gemini-3-pro-image-preview │
│ Iterations: 3

I want to change VLM to openAI

Thanks

[pratikrath126]

Hey! The .env needs two things to use OpenAI instead of Gemini:

1. Uncomment and set VLM_PROVIDER=openai (this controls the main model).
2. (Optional) Set IMAGE_PROVIDER=openai_imagen if you want OpenAI for image generation too.
3. Also make sure OPENAI_API_KEY is filled with your key.

If you only added the API key but left VLM_PROVIDER commented, it will default to Gemini. After editing .env, restart your app. That should switch it over.

[lsdi2016]

@pratikrath126 Thank you for the prompt response. Your inputshelped me in switching the VLM.

Thanks