[Bug] Azure OpenAI o1-mini not support?

[YizhixiaoMoJun]

📦 Environment

Docker

📌 Version

v1.49.12

💻 Operating System

Ubuntu

🌐 Browser

Chrome

🐛 Bug Description

connect to azure openai o1-mini model failed with this error, while others model (4o, 4o-mini) could work as expected.
the configuration looks like: AZURE_MODEL_LIST: gpt-4o->gpt-4o=gpt-4o,gpt-4o-mini->gpt-4o-mini=gpt-4o-mini,o1-mini->o1-mini=o1-mini

{
  "error": {
    "message": "Unsupported value: 'messages[0].role' does not support 'system' with this model.",
    "type": "invalid_request_error",
    "param": "messages[0].role",
    "code": "unsupported_value"
  },
  "endpoint": "https://***.openai.azure.com/",
  "provider": "azure"
}

📷 Recurrence Steps

No response

🚦 Expected Behavior

No response

📝 Additional Information

No response

[lobehubbot]

👀 @YizhixiaoMoJun

Thank you for raising an issue. We will investigate into the matter and get back to you as soon as possible.
Please make sure you have given us as much context as possible.
非常感谢您提交 issue。我们会尽快调查此事，并尽快回复您。 请确保您已经提供了尽可能多的背景信息。

[dosubot]

Hey @YizhixiaoMoJun! I'm here to assist with bugs, questions, and contributions. Let's tackle this together while we wait for a human maintainer.

The Azure OpenAI o1-mini model does not support the 'system' role in the configuration. Instead, it requires the system role message to use "role": "developer". Using the standard system role results in an "unsupported value" error ^[1].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[Sen-CaPoo]

請更新至版本 v1.50.2
#5714

[wolph]

As a workaround you can use litellm to create a proxy that behaves like OpenAI. That way LobeChat will think it is talking to OpenAI o1/o3 (mini) and it will work.

[arvinxx]

@wolph try with the latest version? I think it should be fixed

[wolph]

@wolph try with the latest version? I think it should be fixed

Unfortunately not. I think Lobe-Chat assumes that the model name and the deployment name are identical which is not always the case and in this case I cannot modify it.

I'm not entirely sure what is currently happening, whenever I press generate/regenerate it just appears to be loading for a long time but never generates any results. The logs don't give any output either.

[YizhixiaoMoJun]

Version v1.60.4 has same issue.

{
  "error": {
    "status": 400,
    "headers": {
      "apim-request-id": "8aa27550-8a1f-4a6d-8cb5-342662685a4e",
      "azureml-model-session": "d034-20250212090334",
      "content-length": "221",
      "content-type": "application/json",
      "date": "Mon, 17 Feb 2025 09:52:19 GMT",
      "ms-azureml-model-error-reason": "model_error",
      "ms-azureml-model-error-statuscode": "400",
      "strict-transport-security": "max-age=31536000; includeSubDomains; preload",
      "x-content-type-options": "nosniff",
      "x-envoy-upstream-service-time": "247",
      "x-ms-client-request-id": "8aa27550-8a1f-4a6d-8cb5-342662685a4e",
      "x-ms-rai-invoked": "true",
      "x-ms-region": "East US 2",
      "x-ratelimit-remaining-requests": "63",
      "x-ratelimit-remaining-tokens": "639984",
      "x-request-id": "ebbee39e-7fec-48ae-a503-4bafa8f5d6dc"
    },
    "request_id": "ebbee39e-7fec-48ae-a503-4bafa8f5d6dc",
    "error": {
      "message": "Unsupported value: 'messages[0].role' does not support 'system' with this model.",
      "type": "invalid_request_error",
      "param": "messages[0].role",
      "code": "unsupported_value"
    },
    "code": "unsupported_value",
    "param": "messages[0].role",
    "type": "invalid_request_error"
  },
  "endpoint": "https://***.openai.azure.com/",
  "provider": "azure"
}

[lobehubbot]

✅ @YizhixiaoMoJun

This issue is closed, If you have any questions, you can comment and reply.
此问题已经关闭。如果您有任何问题，可以留言并回复。

[lobehubbot]

🎉 This issue has been resolved in version 1.62.2 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀