NexgAI

curl -X POST https://vllm.nexgai.com/v1/completions \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "What is machine learning?",
    "model": "gpt-oss-20b",
    "max_tokens": 256,
    "temperature": 0.7,
    "persona": "default"
  }'

curl -X POST https://vllm.nexgai.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-oss-20b",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ],
    "max_tokens": 256
  }'

import requests

                            def query_vllm(prompt, model="gpt-oss-20b", persona="default"):
                                response = requests.post(
                                    "https://vllm.nexgai.com/v1/completions",
                                    json={
                                        "prompt": prompt,
                                        "model": model,
                                        "max_tokens": 256,
                                        "temperature": 0.7,
                                        "persona": persona
                                    }
                                )
                                return response.json()

                            # Example usage
                            result = query_vllm("What is AI?")
                            print(result["choices"][0]["text"])

                            # With NBA Analyst persona
                            result = query_vllm(
                                "Analyze LeBron's career stats",
                                persona="nba_analyst"
                            )
                            print(result["choices"][0]["text"])

async function queryVLLM(prompt, options = {}) {
                                const response = await fetch('https://vllm.nexgai.com/v1/completions', {
                                    method: 'POST',
                                    headers: { 'Content-Type': 'application/json' },
                                    body: JSON.stringify({
                                        prompt,
                                        model: options.model || 'gpt-oss-20b',
                                        max_tokens: options.maxTokens || 256,
                                        temperature: options.temperature || 0.7,
                                        persona: options.persona || 'default'
                                    })
                                });
                                return response.json();
                            }

                            // Example usage
                            const result = await queryVLLM('What is machine learning?');
                            console.log(result.choices[0].text);

                            // With options
                            const codeResult = await queryVLLM(
                                'Write a quicksort in Python',
                                { persona: 'code_helper', maxTokens: 512 }
                            );
                            console.log(codeResult.choices[0].text);