ai-gateway
OpenRouter
Unified API for 300+ LLMs across 60+ providers — 1 key, any model
One API for OpenAI, Anthropic, Google, Meta, Mistral, 60+ providers. Auto-fallback, regional routing, pay-as-you-go with 5.5% platform fee.
Pricing
| Tier | Price | Notes |
|---|---|---|
| Free | Free | 25+ free models. 50 requests/day rate limit. 1M free requests/month base. |
| Pay-as-you-go | Free | 5.5% platform fee on usage. Access to 300+ models, 60+ providers. High global rate limits. |
| Enterprise | Custom | Volume-based pricing, bulk discounts, SSO/SAML, dedicated rate limits. 5M free requests/month. |
Limits
| Tier | Metric | Value | Notes |
|---|---|---|---|
| — | auto fallback | If one provider fails, auto-retries with another | Reliability |
| — | billing on success | Only billed for successful runs | Success-only billing |
| — | data retention | Provider data retention disableable per call | Data policy |
| — | no model markup | Model pricing shown in catalog = what you pay (no hidden markup) | Transparent pricing |
| — | openai api compat | Drop-in replacement for OpenAI SDK (change base URL) | API compatibility |
| — | payment methods | Credit card, crypto, bank transfer | Payment options |
| — | platform fee | 5.5% on all usage (Pay-as-you-go) | Platform markup |
| — | regional routing | Pay-as-you-go + Enterprise | Regional options |
Features
- 300+ Models — Claude, GPT, Gemini, Llama, Mistral, Qwen, DeepSeek, Cohere, Grok + open-source. · docs
- 60+ Providers — Anthropic, OpenAI, Google, Together, Fireworks, Groq, DeepInfra, Replicate, etc.
- Auto Fallback — Automatic retry to backup provider on failure.
- Bring Your Own Key — Use your own provider keys → pay providers directly + no platform fee.
- Credit System — Prepay credits via card, crypto, or bank.
- Data Retention Controls — Opt-out of training/retention per provider.
- Free Models Tier — 25+ models available at $0 (limited rate).
- Prompt Caching — Automatic cache for identical prefixes (provider-dependent).
- Provider Preferences — Pin preferred providers per request or default.
- Rankings & Stats — Public leaderboard of most-used models. · docs
- Regional Routing — Route requests to specific geographic regions.
- Streaming — SSE + partial completions.
- Structured Outputs — JSON-mode + JSON schema across supporting models.
- Tool Use / Function Calling — Unified tool calling across providers.
- Unified OpenAI-Compat API — Same endpoint for every model + provider.
Developer interfaces
| Slug | Name | Kind | Version |
|---|---|---|---|
| sdk-openai-compat | Any OpenAI SDK | sdk | — |
| rest-api | OpenRouter API (OpenAI-compat) | rest | v1 |
| dashboard | OpenRouter Dashboard | other | — |
| mcp | OpenRouter MCP | mcp | — |
Compare OpenRouter with
ai-api
OpenRouter vs Anthropic API
Side-by-side breakdown.
ai-api
OpenRouter vs AssemblyAI
Side-by-side breakdown.
ai-api
OpenRouter vs Deepgram
Side-by-side breakdown.
ai-api
OpenRouter vs ElevenLabs
Side-by-side breakdown.
ai-api
OpenRouter vs Google Gemini API
Side-by-side breakdown.
ai-api
OpenRouter vs Groq
Side-by-side breakdown.
ai-api
OpenRouter vs OpenAI API
Side-by-side breakdown.
ai-api
OpenRouter vs Replicate
Side-by-side breakdown.
Staxly is an independent catalog of developer platforms. Outbound links to OpenRouter are plain references to their official pages. Pricing is verified at publication time — reconfirm on the vendor site before buying.