Welcome to OpenRouter. Let's get started

Create your API key and make your first request in minutes. One key for every major model.

OpenRouter ( openrouter.ai?utm_medium=email&utm_source=customerio )

Hi Russell,

*************************************
The Unified Interface For Every Model
*************************************

Welcome to OpenRouter. One API key reaches every major model, with better prices and better uptime: every request routes to the best available provider, with automatic failover when one goes down. Make your first call below.

Text, image, audio, and video models all ship through the same interface: hundreds of models across every top provider, with new ones landing the day they release. Browse the full catalog, filtered by modality, context length, and price.

See every model ( openrouter.ai/models?utm_campaign=onbo…customerio )

Get set up

1. Create your API key ( openrouter.ai/settings/keys?utm_campai…customerio )

One key reaches every model and provider. Scope keys per project, with per-key spend limits.

2. Make your first call ( openrouter.ai/docs/quickstart?utm_camp…customerio )

Paste the snippet below with your key. The API is fully OpenAI compatible, and the response routes to the best live provider automatically.

3. Add credits when you are ready ( openrouter.ai/settings/credits?utm_cam…customerio )

The free tier works without a card. Credits raise your rate limits and open every model.

Paste and run

curl openrouter.ai ( openrouter.ai?utm_medium=email&utm_source=customerio )/api/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer $OPENROUTER_API_KEY" \ -d '{
"model": "~openai/gpt-latest",
"messages": [
{
"role": "user",
"content": "What is the meaning of life?"
}
]
}'

The ~ prefix is a latest alias: it always resolves to the newest flagship in that family. Swap the model field for any slug on rankings ( openrouter.ai/rankings?utm_campaign=on…customerio ) and the same call keeps working.

-------------------
Models and rankings
-------------------

Every model is ranked by real production traffic, not marketing. The rankings ( openrouter.ai/rankings?utm_campaign=on…customerio ) show what teams actually run in production, updated as usage shifts between models. It is the fastest read on what is working right now.

Featured

anthropic/claude-opus-5 ( openrouter.ai/anthropic/claude-opus-5?…customerio )
Anthropic’s flagship for demanding reasoning, coding, and long-horizon agentic work. Text, image, and file input, 1M context.

google/gemini-3.7-flash ( openrouter.ai/google/gemini-3.7-flash?…customerio )
fast multimodal: text, image, audio, and video input, 1M context

openrouter/auto ( openrouter.ai/openrouter/auto?utm_camp…customerio )
routes each request to the best model for the task

See the rankings ( openrouter.ai/rankings?utm_campaign=on…customerio )

**********************************
New models are live on OpenRouter.
**********************************

New models land on OpenRouter constantly, across text, image, audio, and video. One API key reaches all of them, and your code stays the same when you swap the model slug. Here is what is new right now.

Featured

z-ai/glm-5.3 ( openrouter.ai/models/z-ai/glm-5.3?utm_…customerio )
GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.

Also new on OpenRouter

qwen/qwen3.8-27b ( openrouter.ai/models/qwen/qwen3.8-27b?…customerio )text, image, video

google/gemini-3.7-flash ( openrouter.ai/models/google/gemini-3.7…customerio )text, image, video, file

bytedance-seed/seed-2-1-turbo ( openrouter.ai/models/bytedance-seed/se…customerio )text, image, video

qwen/qwen3.8-2.4t-a95b ( openrouter.ai/models/qwen/qwen3.8-2.4t…customerio )text

bytedance-seed/seed-2.0-code ( openrouter.ai/models/bytedance-seed/se…customerio )text, image, video

deepseek/deepseek-v4-pro-0813 ( openrouter.ai/models/deepseek/deepseek…customerio )text

See all new models ( openrouter.ai/models?sort=newest&utm_m…customerio )

----------
Benchmarks
----------

The benchmarks ( openrouter.ai/benchmarks?utm_campaign=…customerio ) are independent, reproducible measurements of the knobs you can set on a request: models, providers, search engines, and tool budgets, scored for quality, value, and speed. Every score links to the configuration, cost, and telemetry behind it.

See the benchmarks ( openrouter.ai/benchmarks?utm_campaign=…customerio )

---------------
Built for teams
---------------

Workspaces scope keys, spend limits, and provider rules per project; guardrails set budget caps and policy controls across every key. Zero-data-retention routing is a per-request flag, and EU in-region processing runs on a dedicated base URL for enterprise accounts. If you are evaluating gateways for an org, talk to the enterprise team ( openrouter.ai/enterprise?utm_campaign=…customerio ).

Set up your workspace ( openrouter.ai/docs/guides/overview/aut…customerio )

Latest from the blog

OpenRouter is Joining Stripe ( openrouter.ai/announcements/openrouter…customerio )

OpenRouter is Joining Stripe ( openrouter.ai/announcements/openrouter…customerio )
Today, we are excited to announce that we are joining forces with Stripe, to power the next wave of GDP growth globally.

Read the announcement → ( openrouter.ai/announcements/openrouter…customerio )

Understand your AI usage: every agent, model, and request ( openrouter.ai/announcements/activity-d…customerio )

Understand your AI usage: every agent, model, and request ( openrouter.ai/announcements/activity-d…customerio )
See what your team spent on every model, save the charts you keep rebuilding, click any bar to land in the logs behind it, and query the same data from your terminal with the Analytics API.

Read the announcement → ( openrouter.ai/announcements/activity-d…customerio )

Ori Eval: Find the Best Model for What You're Building ( openrouter.ai/announcements/ori-eval?u…customerio )

Ori Eval: Find the Best Model for What You're Building ( openrouter.ai/announcements/ori-eval?u…customerio )
Choosing the model inside your product is often decided without a systematic method. Ori Eval answers the question with proof: it runs your agent on your own prompts, checks the tools it called, and grades the answers.

Read the announcement → ( openrouter.ai/announcements/ori-eval?u…customerio )

We are always shipping. Keep up with our latest insights, announcements, and more on our blog → ( openrouter.ai/blog?utm_medium=email&ut…customerio )

openrouter.ai ( openrouter.ai?utm_medium=email&utm_source=customerio )·Discord ( discord.gg/openrouter?utm_medium=email…customerio )·X ( x.com/openrouter?utm_medium=email&utm_…customerio )·Docs ( openrouter.ai/docs?utm_medium=email&ut…customerio )

169 Madison Avenue, New York, NY 10016

Manage preferences ( track.customer.io/manage_subscription_…jDTjdXYQ== )|Unsubscribe ( track.customer.io/unsubscribe/dgSE_QoB…jDTjdXYQ== )