Unable to connect to AzureOpneAI

[shwetabelt]

I have correctly configured my API key, base URL, API version, and deployment name in the .env file. I’ve also verified the setup using curl, and it works as expected.

However, when I send any prompt through LibreChat, I consistently receive the following error:

"Something went wrong. Here's the specific error message we encountered: An error occurred while processing the request: 404 Resource not found Troubleshooting URL: https://js.langchain.com/docs/troubleshooting/errors/MODEL_NOT_FOUND/"

For context, I'm using Azure OpenAI with the gpt-4o-mini model.

I have tried using the Gemini model as well but this is what i got with different gemini models.

"error: [api/server/controllers/agents/client.js #sendCompletion] Operation aborted [GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-pro-001:streamGenerateContent?alt=sse: [404 Not Found] models/gemini-1.5-pro-001 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."

Other providers like Anthropic and Ollama are working without any issues.

Also there is another issue when i am trying to upload a text file, I have provided the OCR key and rag_api is also up and running, still its failing with following error:

"2025-07-14 15:31:09 2025-07-14 10:01:09 error: Not Found The server responded with status 404: Request failed with status code 404
2025-07-14 15:31:09 2025-07-14 10:01:09 error: [/files] Error processing file: Not Found The server responded with status 404: Request failed with status code 404"

[samuelhalff]

You will probably have to share your config and logs help I'm afraid.

[danny-avila]

Please review: https://www.librechat.ai/docs/configuration/azure

[shwetabelt]

This is my librechat.yaml

version: 1.2.8 cache: true endpoints: azureOpenAI: streamRate: 35 titleModel: "llama-70b-chat" titleConvo: true assistants: true plugins: true groups: - group: "eastus" assistants: true apiKey: "${AZURE_OPENAI_API_KEY}" version: "2025-01-01-preview" baseURL: "https://xxxx.openai.azure.com/openai/deployments/gpt-4.1-mini" instanceName: "gpt-4.1-mini" models: gpt-4.1-mini: deploymentName: "gpt-4.1-mini" version: "2025-01-01-preview"

With this i was able to run AzureOpenAI and its giving response as well. But now i am facing 2 problems.

1. /api/convos/gen_title api is breaking and giving the following response :
   {
   "message": "Title not found or method not implemented for the conversation's endpoint"
   }
   I even took the latest pull as i saw some comments saying this wasn't working 2 days ago but its still not working for me.

2. Currently i have done the setup on local but i am planning to deploy it on our host. But before that i wanted to ask if i can use the apis directly to pass the context (documents + Some email data), that i want to be pre-populated with the selected model by default as this is my requirement. Is it doable? Does Librechat currently supports it? If yes, then can you please provide me some documentation?

Also the code seems to be constantly changing, so even if we host it and starts using it, with continuous code change there will be maintenance challenge. How are you currently taking care of it?

[shwetabelt]

@danny-avila Please let me know if its possible to pre-populate or load the context and file via apis directly.