Trường hợp sử dụng

import OpenModex from '@openmodex/sdk';

const client = new OpenModex({
  apiKey: process.env.OPENMODEX_API_KEY,
  fallbackModels: ['claude-3.5-sonnet', 'gemini-2.0-flash'],
});

// Use GPT-4o for complex questions, mini for simple ones
const stream = await client.chat.completions.create({
  model: isComplexQuery ? 'gpt-4o' : 'gpt-4o-mini',
  messages: conversationHistory,
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}

// 1. Create embeddings for your knowledge base
const embedding = await client.embeddings.create({
  model: 'text-embedding-3-small',
  input: userQuery,
});

// 2. Search your vector database
const relevantDocs = await vectorDB.search(embedding.data[0].embedding);

// 3. Generate an answer with context
const response = await client.chat.completions.create({
  model: 'gpt-4o',
  messages: [
    { role: 'system', content: `Answer based on this context:\n${relevantDocs}` },
    { role: 'user', content: userQuery },
  ],
  routing: { strategy: 'quality_optimized' },
});

// Generate 1000 product descriptions cost-efficiently
const descriptions = await Promise.all(
  products.map(async (product) => {
    const response = await client.chat.completions.create({
      model: 'gpt-4o-mini', // cost-effective for structured tasks
      messages: [
        { role: 'system', content: 'Write a 100-word product description.' },
        { role: 'user', content: JSON.stringify(product) },
      ],
      routing: { strategy: 'cost_optimized' },
      cache: { enabled: true, ttl: 86400 }, // cache for 24h
    });
    return response.choices[0].message.content;
  })
);

const response = await client.chat.completions.create({
  model: 'claude-3.5-sonnet', // best at code
  messages: [
    { role: 'system', content: 'You are a senior software engineer.' },
    { role: 'user', content: `Review this code and suggest improvements:\n${code}` },
  ],
  routing: {
    fallback: ['deepseek-chat', 'gpt-4o'], // code-capable fallbacks
  },
});

// Compare responses from different models
const comparison = await client.models.compare([
  'openai/gpt-4o',
  'anthropic/claude-3.5-sonnet',
  'google/gemini-2.0-flash',
]);

// Run the same prompt against multiple models
const results = await Promise.all(
  ['gpt-4o', 'claude-3.5-sonnet', 'gemini-2.0-flash'].map(async (model) => {
    const response = await client.chat.completions.create({
      model,
      messages: [{ role: 'user', content: testPrompt }],
    });
    return {
      model,
      content: response.choices[0].message.content,
      cost: response.usage?.cost_usd,
      latency: response.openmodex?.latency_ms,
    };
  })
);

import { useChat } from '@openmodex/react';

function SupportChat() {
  const { messages, input, setInput, sendMessage, isLoading, stop } = useChat({
    apiKey: process.env.NEXT_PUBLIC_OPENMODEX_KEY,
    model: 'gpt-4o-mini',
    system: 'You are a helpful customer support agent for Acme Corp.',
  });

  return (
    <div className="chat-container">
      {messages.map((m) => (
        <div key={m.id} className={`message ${m.role}`}>
          {m.content}
        </div>
      ))}
      <form onSubmit={(e) => { e.preventDefault(); sendMessage(); }}>
        <input value={input} onChange={(e) => setInput(e.target.value)} />
        <button disabled={isLoading}>Send</button>
      </form>
    </div>
  );
}