NEW: Claude Fable 5, Gemini 3.5 Flash & GPT-5.6 Sol Pro Now Live

Every frontier model.
One API.

Call Claude, Gemini, GPT, DeepSeek, Qwen and more from a single OpenAI-compatible endpoint. Build agents and RAG with web-extract built in — free to start, no credit card.

Get Your Free API Key View documentation
API key in 30 seconds OpenAI-compatible Free API key, no credit card Copy-paste quickstart
Terminal
curl https://api.zyloai.net/v1/chat/completions \
  -H "Authorization: Bearer YOUR_ZYLO_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.5",
    "messages": [{"role": "user", "content": "Hello from Zylo!"}]
  }'
Python
# pip install openai
from openai import OpenAI

client = OpenAI(api_key="YOUR_ZYLO_KEY", base_url="https://api.zyloai.net/v1")

response = client.chat.completions.create(
    model="gpt-5.5",
    messages=[{"role": "user", "content": "Hello from Zylo!"}],
)
print(response.choices[0].message.content)
Node.js
// npm install openai
import OpenAI from "openai";

const client = new OpenAI({ apiKey: "YOUR_ZYLO_KEY", baseURL: "https://api.zyloai.net/v1" });

const response = await client.chat.completions.create({
  model: "gpt-5.5",
  messages: [{ role: "user", content: "Hello from Zylo!" }],
});
console.log(response.choices[0].message.content);

Why choose us?

The developer-first platform designed for speed, scale, and simplicity.

Unified Integration

Access 40+ state-of-the-art models from OpenAI, Anthropic, Google, DeepSeek, Qwen and more with a single OpenAI-compatible SDK.

Ultra-Low Latency

Our global edge network ensures your requests are routed to the fastest available instance with millisecond precision.

Zero Data Retention

Your data stays yours. We offer zero-retention policies on all premium models, ensuring complete privacy.

Infinite Scalability

From prototype to production, our infrastructure handles millions of tokens per minute without breaking a sweat.

Supported Models

Industry-leading performance across the world's best architectures.

Claude
Claude Fable 5
Anthropic
Gemini
Gemini 3.5 Flash
Google
GPT
GPT-5.6 Sol Pro
OpenAI
DeepSeek
DeepSeek V4 Pro
DeepSeek
MiniMax
MiniMax M3
MiniMax
Qwen
Qwen 3.7 Max
Alibaba
Kimi
Kimi K2.6
Moonshot AI

Zylo vs OpenRouter vs going direct

All three let you call large language models over HTTP. Here is how Zylo's gateway compares with OpenRouter and integrating each provider directly.

Capability Zylo OpenRouter Direct providers
Many providers, one key Yes — 7 providers Yes — very wide No — one each
OpenAI-compatible endpoint Yes Yes Varies by provider
Usage pricing Base per-token rate, no markup Pay-as-you-go Each provider's list price
Platform fee 25% only when you add credits Varies — check current terms None / provider-set
Built-in web-extract for RAG Yes No No
Local payment methods Yes Varies Varies
Free tier to start Yes — Basic, no card Varies Varies
Competitor offerings change frequently — verify current terms with each provider. Last reviewed June 2026. See the deeper breakdown in Zylo vs OpenRouter.

Simple Monthly Plans

Transparent pricing that scales with your ambition.

Basic

$0/mo
  • 7.2k req/day · 10/min
  • 200k Daily Tokens
  • Basic models only
  • Max 200k input chars/req
  • Text only (no multimodal)
  • No credits
Get Started

Go

$10/mo
  • 28.8k Daily Req
  • 512k Daily Tokens
  • Web Search
  • $10 in credits
Upgrade now

Pro

$50/mo
  • 43.2k Daily Req
  • 1M Daily Tokens
  • Code Execution
  • $50 in credits
Go Pro

Mega

$200/mo
  • 86.4k Daily Req
  • 5M Daily Tokens
  • Priority Access
  • $200 in credits
Upgrade now

Enterprise

$400/mo
  • Unlimited Req
  • Unlimited Tokens
  • Dedicated GPU
  • $400 in credits
Get access

Flexible Budgeting

Estimate your tokens for custom credit purchases.

$50
Claude Fable 5 --
Gemini 3.5 Flash --
GPT-5.6 Sol Pro --
Open Models --
Purchase Credits

Popular guides

Comparisons, migrations and how-tos for building on one OpenAI-compatible API.

Comparison

Zylo vs OpenRouter

Pricing, models, retrieval and payments compared — and how to migrate in two lines.

Read article → Comparison

The cheapest LLM APIs in 2026

A price-per-million comparison of the cheapest models you can call today, plus tips to cut your bill.

Read article → Guide

Best AI API gateway for developers

A practical checklist for choosing a gateway — and how Zylo measures up.

Read article → Migration

Coming from OpenRouter?

Repoint the base URL, use bare model ids, and bring your code over in minutes.

Read guide → Migration

Switch from OpenAI in 30 seconds

Keep your OpenAI SDK; change two lines and pick any model.

Read guide → Blog

All guides & comparisons

Browse every Zylo guide — routing, RAG with web-extract, coding agents and more.

Visit the blog →

Explore Zylo

Browse by provider, compare models head-to-head, or see how Zylo stacks up as an alternative.

All models & pricing By provider Compare models Alternatives Code agents Pricing Cost calculator

Frequently asked questions

The essentials on compatibility, pricing and the Basic plan.

What is Zylo AI?

Zylo AI is an OpenAI-compatible API that gives you one key and one base URL to call frontier models from Anthropic, OpenAI, Google, DeepSeek and more. Point any OpenAI SDK at https://api.zyloai.net/v1, change the model id, and you are done.

Is Zylo OpenAI-compatible?

Yes. Keep the official OpenAI SDK and your code; set base_url to https://api.zyloai.net/v1 and use your Zylo API key. Chat completions, streaming and tool calling all work unchanged.

How does Zylo's pricing work?

Each model is billed at its base per-token rate, deducted from your prepaid credit balance — there is no markup on usage. Zylo's flat 25% platform fee applies only when you add credits, never on what you spend.

Is the Basic plan free, and are all models included?

Yes — the Basic plan is free with no credit card. It gives a daily usage allowance and a global limit of 10 requests per minute, and includes Basic-tier models only (not premium models). Basic does not include credits; paid plans (Go and up) add premium models, higher limits and credits you spend on usage.

Which AI models does Zylo support?

Frontier and cost-efficient models from seven providers — Anthropic (Claude), OpenAI (GPT), Google (Gemini), DeepSeek, Qwen, MiniMax and Moonshot (Kimi). See the models page for the live list and pricing.

How is Zylo different from OpenRouter?

Zylo focuses on transparent base pricing (the 25% fee applies only at credit top-up, not on usage), built-in web-extract for RAG, local payment methods, and bare model ids like claude-opus-4.8 instead of vendor/model strings.

Terms of Service

Last updated: February 22, 2026

1. API Requests Policy

All requests are subject to automated safety and abuse detection.

2. Prohibited Activities

The following activities and content generation are strictly prohibited:

3. Linked Accounts

Accounts may be considered linked if they demonstrate highly correlated usage patterns, technical signals, or behaviors that indicate shared control, coordination, or attempts to bypass usage limits.

We do not require accounts to be owned by the same individual to be considered linked.

4. Violations and Enforcement

Violations may result in actions including, but not limited to: rate limiting, quota reduction, account restriction, API key suspension, or termination of service.

Back to Home

Privacy Policy

Last updated: June 2, 2026

1. Authentication & Google Login

Our platform uses Google OAuth for secure authentication. We only request basic profile information (like your email address) to create and identify your account. We do not have access to your Google password or any other private information from your Google account.

2. Account Data Storage

We store your account configuration, API key, plan, and credit balance in a managed PostgreSQL database (Supabase), encrypted at rest. Access is enforced by row-level security, so each account can only read its own data, and sensitive fields such as your plan and credit balance can only be modified by our secured backend — never directly from your browser.

3. Zero Data Retention Policy

We value your privacy. Your inputs, prompts, and the generated AI completions are never stored on our servers. We operate as a high-speed passthrough to the model providers. The content of your requests vanishes the moment the request is completed.

4. What We Do Log

The only data we actively log and monitor is numerical usage data. We record the number of tokens used, the model requested, and the timestamp of the request. This is strictly required for billing, rate-limiting, and preventing abuse of the API.

Back to Home