OpenRouterNew

One API key for hundreds of models, with provider prices passed through, a percentage fee on credits, routing fallbacks and a per-request price cap.

  • Popularity
  • Model API
  • Multi-model
  • Routing
  • Developer Tools
  • Coding
  • Free tier
No preview available

At a glance

One key, one balance, hundreds of models, with automatic provider selection by price and latency — the least tedious way to evaluate models.

  • Free tierPartialFree models with rate limits; real usage needs credits
  • Open sourceNo
  • Works in ChinaNo
  • APIYes
Best for
  • Developers comparing several models side by side
  • Products that need automatic fallback when a provider fails
  • Teams who don't want separate onboarding and billing per vendor
Pros
  • OpenAI-compatible endpoint, so integration barely changes your code
  • Provider token prices are passed through; the fee lands on credit purchases
  • Routing variants, a per-request price ceiling and custom data policies
Cons
  • A single-provider product with no fallback need pays routing overhead for nothing
  • If a provider raises rates, requests still route there and bill at the new price
  • Public SLA terms are limited; heavy production use means the enterprise plan
Pricing

A free tier offers 25+ free models with rate and daily limits; pay-as-you-go uses prepaid credits with pass-through token pricing and a percentage fee on purchases; enterprise adds SSO and SLAs. Check the site for current rates.

Pricing changes over time; check the official site

Access from mainland China

Reaching openrouter.ai from mainland China requires your own network arrangements.

OpenRouter is an aggregation layer for model APIs: one key, one balance, and access to hundreds of models from dozens of providers through an OpenAI-compatible endpoint, with routing, price comparison and failover between them.

The Problem It Removes

When you evaluate several models at once, the time sink isn't the code — it's repeating onboarding, key management, billing and error handling four times over. OpenRouter collapses that into one entry point: switching models is usually a change to one model-name string.

For products already in production, the value sits at the other end — automatic fallback when a provider degrades, without waiting for a status page to update.

How Billing Works

Token prices are passed through at each provider's own rate, with the platform's fee applied when you buy credits. That keeps per-model prices in line with going direct and makes the overhead knowable in advance. The free tier includes 25+ free models with rate and daily request caps — enough to exercise routing logic and compare behavior. Bring-your-own-key is supported too, with a small usage fee once you pass the free monthly allowance.

Knobs Worth Knowing

  • Routing variants: pick providers dynamically by cost or performance.
  • Per-request price ceiling: set a threshold, and the request isn't sent if no provider meets it.
  • Data policies: restrict routing to providers you trust; prompts and completions aren't logged by default.

When Not to Use It

If your product uses one model and needs no fallback, this layer is pure overhead — go direct. Also note that when a provider changes rates, requests keep routing there and bill at the new price, so pair budget-sensitive workloads with the price cap.

Using It From China

Reaching openrouter.ai from mainland China requires your own network arrangements.