OkeyMeta AI Platform

Build Intelligent Applications Create Custom AI Models Process Natural Language Analyze Images and Data Generate Code Automatically

Powered by SSAILM 3.0
Nigeria's First Enterprise-Grade AI Platform

20T
250+
50ms

Processing your request...

API Documentation

Base URL

https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai

Chat Endpoint

Send messages to OkeyAI and receive responses

Query Parameters

Parameter Type Required Description
input string Yes The message to send to OkeyAI
imgUrl string No URL of an image to analyze
fineTunedAccessKey string No Access key for using a fine-tuned model

Example Request

curl "https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai?input=Tell%20me%20about%20Nigeria"

Example Response

{
    "status": "200",
    "model": {
        "name": "OkeyAI 3.0 Vanguard",
        "developer": "Okechukwu Nwaozor",
        "team": "OkeyMeta Programming Team"
    },
    "description": "A cutting-edge AI that sees, hears, understands, and adapts in real time.",
    "response": "Nigeria is the most populous country in Africa...",
    "token_count": 245,
    "API_KEY": "okeymeta-abc123...",
    "metrics": {
        "response_time": "1.2 seconds",
        "accuracy": 0.95,
        "confidence_score": "high"
    }
}

Fine-tuning Endpoint

Create a customized version of OkeyAI

Query Parameters

Parameter Type Required Description
mode string Yes Set to "finetune"
modelName string Yes Name for your fine-tuned model
version string Yes Version identifier
developer string Yes Developer name
tone string Yes Model's tone (professional, casual, etc.)
trainingData string Yes JSON array of training examples

Fine-tuning API Documentation

Create a customized version of OkeyAI with specific behaviors and knowledge

Endpoint

GET ${API_BASE_URL}?mode=finetune

Query Parameters

Parameter Type Required Description
mode string Yes Must be set to "finetune"
modelName string Yes Unique identifier for your model
version string Yes Version number (e.g., "1.0.0")
developer string Yes Name of the developer or organization
specialization string Yes Model's area of expertise
tone string Yes One of: professional, casual, friendly, technical, academic, medical, legal, business
additionalInstructions string No Special instructions for the model
trainingData JSON Yes Training examples and conversations

Example Request

curl "${API_BASE_URL}?mode=finetune&modelName=MedicalAssistant&version=1.0.0&developer=DrJohn&specialization=Medical%20Diagnosis&tone=professional&trainingData=%7B%22conversations%22:%5B%7B%22input%22:%22What%20are%20the%20symptoms%22,%22output%22:%22Let%20me%20help%20diagnose%22%7D%5D%7D"

Example Response

{
    "status": "200",
    "message": "Fine-tuned model created successfully",
    "accessKey": "okeymeta-abc123...",
    "modelDetails": {
        "name": "MedicalAssistant",
        "version": "1.0.0",
        "developer": "DrJohn",
        "specialization": "Medical Diagnosis",
        "tone": "professional",
        "baseModel": "OkeyAI 3.0 Vanguard"
    }
}

Model Update API Documentation

Update an existing fine-tuned model's parameters

Endpoint

GET ${API_BASE_URL}?mode=update

Query Parameters

Parameter Type Required Description
mode string Yes Must be set to "update"
accessKey string Yes The model's access key
modelName string No New model name
version string No New version number
tone string No New communication tone
specialization string No Updated specialization

Conversational Memory

OkeyAI maintains conversation history for each unique API key. To enable this:

  • First request: Leave API key empty to get a new key in the response
  • Subsequent requests: Include the same API key to continue the conversation
  • Each user should have their own unique API key
  • Conversations are maintained for up to 50 interactions

Example Flow

// First request (no API key)
GET ${API_BASE_URL}?input=Hello

// Response includes new API key
{
    "API_KEY": "okeymeta-abc123...",
    "response": "Hello! How can I help you?"
}

// Subsequent request (with API key)
GET ${API_BASE_URL}?input=Tell me more&APiKey=okeymeta-abc123...

Code Examples


curl -X GET "https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai?input=Hello" \
    -H "Accept: application/json"

import requests

def chat_with_okeymeta(message, api_key=None):
    url = "https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai"
    params = {
        "input": message
    }
    if (api_key) {
        params["APiKey"] = api_key;
    }

    response = requests.get(url, params=params)
    return response.json()

# Example usage
response = chat_with_okeymeta("Tell me about Nigeria")
print(response["response"])
                            

async function chatWithOkeyMeta(message, APiKey = null) {
    const url = new URL('https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai');
    url.searchParams.append('input', message);
    if (APiKey) {
        url.searchParams.append('APiKey', APiKey);
    }

    const response = await fetch(url);
    const data = await response.json();
    return data;
}

// Example usage
chatWithOkeyMeta("Tell me about Nigeria")
    .then(data => console.log(data.response))
    .catch(error => console.error(error));
                            

import java.net.URI;
import java.net.http.*;

public class OkeyMetaAI {
    private static final String BASE_URL = 
        "https://api.okeymeta.com.ng/api/ssailm/model/okeyai3.0-vanguard/okeyai";
    
    public static String chat(String message, String APiKey) throws Exception {
        String url = BASE_URL + "?input=" + URLEncoder.encode(message, "UTF-8");
        if (APiKey != null) {
            url += "&APiKey=" + URLEncoder.encode(APiKey, "UTF-8");
        }
        
        HttpClient client = HttpClient.newHttpClient();
        HttpRequest request = HttpRequest.newBuilder()
            .uri(URI.create(url))
            .GET()
            .build();
            
        HttpResponse response = client.send(
            request, 
            HttpResponse.BodyHandlers.ofString()
        );
        
        return response.body();
    }
}