gpt-image-1 on Azure

[gitanna]

I deployed gpt-image-1 on Azure and tested it via command line: curl -X POST "https:/--westus3.cognitiveservices.azure.com/openai/deployments/gpt-image-1/images/generations?api-version=2025-04-01-preview" -H "Content-Type: application/json" -H "api-key:" -d '{"prompt": "test image", "n": 1, "size": "1024x1024"}' and it works fine. But if accessed through the menu, there is an error: "Making request to https://--westus3.cognitiveservices.azure.com/openai/deployments/gpt-image-1/chat/completions
warn: [OpenAIClient.chatCompletion][stream] API error
error: [AskController] Error handling request 404 Resource not found" "https://-****-westus3.cognitiveservices.azure.com/openai/deployments/gpt-image-1/" , it treats is as chat completion.

Here is what I have in .env:
`IMAGE_GEN_OAI_API_KEY=

IMAGE_GEN_OAI_BASEURL=https://-*-westus3.openai.azure.com/openai/deployments/gpt-image-1/

IMAGE_GEN_OAI_AZURE_API_VERSION=2025-04-01-preview
`
This is my librechat.yaml:

`custom:

 - name: "GPT Image"
 
    apiKey: <key>

    baseURL: "https://****-****-westus3.cognitiveservices.azure.com/openai/deployments/gpt-image-1/images/generations?api-version=2025-04-01-preview"

    models:

        default: [

          "gpt-image-1"

        ]

     modelDisplayLabel: "GPT-image-1"`

[kamaldua65]

You can not access it directly, you need to make an agent. Use any chat model when creating agent and then add tool gp-image-1.

[gitanna]

Thank you. It worked.

[kjanssens-ili]

@gitanna Did this work for you with the Azure API key? For me that returns a 401, telling me i need an OpenAI API key.

[aekusal]

Can you please share the librechat.yaml config you used to get this to work? @gitanna I have tried everything and am stuck on this.