We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
LLM API for chat, agents and structured output
Use one chat completions endpoint for every large language model. Streaming, tool calling, JSON schema outputs, vision and reasoning work the same way across providers, so you can pick the best price or quality per task without rewriting code.
- Chatbots and support assistants
- Agents that call your tools and APIs
- Extracting structured JSON from documents
- Summaries, translation and classification at scale
One API. Every model.
Session cost $0.00
Settings
Advanced parameters
LLM API · 452 models
How to call the LLM API API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "qwen/qwen3.6-27b".
How do I choose an LLM?
Start with a fast, low-cost model for routing and extraction, and reserve frontier models for complex reasoning. The table below is sorted by price so you can compare quickly.
Can I switch models without code changes?
Yes. Model ids use the provider/model format (for example openai/gpt-5-mini); change the model string in the same request.
How do I keep LLM costs down?
Cap max_tokens, reuse long system prompts with prompt caching, and send simple tasks to a small model. Each model page shows the cost of a typical request.
Does streaming cost more?
No. Streamed and non-streamed responses are billed the same way, by input and output tokens.
Which models support tool calling and JSON output?
Each model page lists its supported parameters. Look for tools and structured outputs, or filter the catalog by capability.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "qwen/qwen3.6-27b",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);