Answer API

import { LLMLayerClient } from 'llmlayer';

const client = new LLMLayerClient({ apiKey: process.env.LLMLAYER_API_KEY! });

// Basic answer
const response = await client.answer({
query: 'What are the latest AI breakthroughs?',
model: 'openai/gpt-4o-mini'
});

console.log(response.llm_response); // Markdown by default

const resp = await client.answer({
query: 'Explain quantum computing in simple terms',
model: 'openai/gpt-4o-mini',
temperature: 0.7,
maxTokens: 1000,
returnSources: true,
});

console.log(resp.llm_response);
const total = (resp.model_cost ?? 0) + (resp.llmlayer_cost ?? 0);
console.log(`Total cost: $${total.toFixed(4)}`);

const schema = {
type: 'object',
properties: {
topic: { type: 'string' },
key_concepts: { type: 'array', items: { type: 'string' } },
applications: {
type: 'array',
items: {
type: 'object',
properties: { name: { type: 'string' }, description: { type: 'string' } },
required: ['name', 'description']
}
}
},
required: ['topic', 'key_concepts', 'applications']
};

const resp = await client.answer({
query: 'Applications of machine learning in healthcare',
model: 'openai/gpt-4o',
answerType: 'json',
jsonSchema: schema,
maxQueries: 2,
searchContextSize: 'high',
});

const data = typeof resp.llm_response === 'string' ? JSON.parse(resp.llm_response) : resp.llm_response;
console.log(data.topic, data.key_concepts.length);

const resp = await client.answer({
query: 'Latest developments in renewable energy',
model: 'anthropic/claude-sonnet-4',
searchType: 'news',
dateFilter: 'week',
citations: true,
returnSources: true,
returnImages: true,
maxQueries: 3,
domainFilter: ['reuters.com', 'bloomberg.com', '-reddit.com'],
systemPrompt: 'Focus on breakthroughs and policy changes',
});
console.log(resp.llm_response);

const resp = await client.answer({
query: 'Complex medical research query',
model: 'openai/gpt-4o',
provider_key: process.env.OPENAI_API_KEY!,
maxTokens: 4000,
domainFilter: ['pubmed.gov', 'nature.com', 'sciencedirect.com'],
searchContextSize: 'high',
});
console.log('LLMLayer cost:', resp.llmlayer_cost); // ~$0.004 × max_queries
console.log('Model cost:', resp.model_cost);       // null when using provider_key

const stream = client.streamAnswer({
query: 'Explain the history of the internet',
model: 'groq/llama-3.3-70b-versatile',
returnSources: true,
temperature: 0.5,
});

for await (const event of stream) {
switch (event.type) {
case 'llm':
process.stdout.write(String(event.content));
break;
case 'sources':
console.log('\nSources:', event.data);
break;
case 'images':
console.log('\nImages:', event.data?.length ?? 0);
break;
case 'usage':
const total = (event.model_cost ?? 0) + (event.llmlayer_cost ?? 0);
console.log('\nTokens:', event.input_tokens);
console.log('Total cost:', total);
break;
case 'done':
console.log('\nCompleted in', event.response_time, 'seconds');
break;
case 'error':
console.error('\nStream error:', event.error);
break;
}
}

import { LLMLayerClient, AuthenticationError, InvalidRequest, RateLimitError, ProviderError } from 'llmlayer';

const client = new LLMLayerClient({ apiKey: process.env.LLMLAYER_API_KEY! });

async function robustAnswer(query: string, retries = 3) {
for (let i = 0; i < retries; i++) {
try {
return await client.answer({ query, model: 'openai/gpt-4o-mini' });
} catch (err) {
if (err instanceof AuthenticationError) throw err; // fix credentials
if (err instanceof RateLimitError) {
await new Promise(r => setTimeout(r, Math.pow(2, i) * 1000));
continue;
}
if (err instanceof ProviderError) {
// fallback model
return await client.answer({ query, model: 'groq/llama-3.3-70b-versatile' });
}
throw err;
}
}
throw new Error('Max retries exceeded');
}