How to handle "This model has no interface set up yet."?

[helix]

Versions (relevant - OpenSearch/Dashboard/Server OS/Browser):3.2.0

Describe the issue:I have completed the test model inference and received the dialogue response from the model. However, after setting up the AI search flow and adding sample data, when I click on the ML Inference Processor, this message appears. Did I miss any configuration?

Configuration:Create connector for DeepSeek Chat，Create model group，Register model to model group & deploy model，Test model inference，I can receive responses from DeepSeek and converse with it.in machine learning，DeepSeek Cha is Responding

Relevant Logs or Screenshots:

[pablo]

@helix I’ve followed OpenSearch documentation and successfully registered OpenAI model with interface.

Register model

This is my working example.

POST /_plugins/_ml/models/_register
{
    "name": "myopenai_interface",
    "function_name": "remote",
    "connector_id": "6lsCdpkBhSMpS9gk7NeY",
    "description": "test model",
    "interface": {
        "input": {
            "properties": {
                "parameters": {
                    "properties": {
                        "messages": {
                            "type": "string",
                            "description": "This is a test description field"
                        }
                    }
                }
            }
        },
        "output": {
            "properties": {
                "inference_results": {
                    "type": "array",
                    "items": {
                        "type": "object",
                        "properties": {
                            "output": {
                                "type": "array",
                                "items": {
                                    "properties": {
                                        "name": {
                                            "type": "string",
                                            "description": "This is a test description field"
                                        },
                                        "dataAsMap": {
                                            "type": "object",
                                            "description": "This is a test description field"
                                        }
                                    }
                                },
                                "description": "This is a test description field"
                            },
                            "status_code": {
                                "type": "integer",
                                "description": "This is a test description field"
                            }
                        }
                    },
                    "description": "This is a test description field"
                }
            }
        }
    }
}

[helix]

Thank you for your answer, I have completed the registration. But I have a new question: What is an Embedding Model? I haven’t found any relevant documentation, and I am not a developer but a self-learner with no prior knowledge. Could you provide me with some relevant clues? Thank you, and I wish you a great day!

[pablo]

@helix In general it means any model that will translate your input (text, picture, video) into a vector (embedding)

You can find some examples by clicking Learn More link at Embedding Models.

documentation/models.md