YoloRouter
yolorouter.com
- Free plan
- Open source
- Windows
- macOS
- Linux
- Web
- API
- Self-hosted
What YoloRouter is
YoloRouter is a gateway for routing AI model requests across multiple providers. It accepts OpenAI Chat Completions and Responses, Anthropic Messages and Gemini generateContent protocols, and translates between supported formats when the client and provider use different ones. Routing includes ordered provider choices, failover and upstream key pooling. Streaming uses Server-Sent Events with usage tracking; the gateway also offers OpenAI Images and Videos APIs, with charges based on delivered images or video seconds. It can compress bulky tool output before forwarding requests and report measured savings. API keys can be limited by model allowlists, rate and concurrency caps, cumulative budgets, expiry and revocation. Its admin console supports multiple users, OAuth2/OIDC single sign-on, usage visibility and account controls. For self-hosting, the product comes as a single binary with an embedded admin console, SQLite by default and optional PostgreSQL. Provider keys are encrypted at rest with AES-256, while admin credentials and API keys are hashed. The self-hosted plan is free; cloud billing uses top-up credits without subscription or minimum spend.
Who it suits
YoloRouter is aimed at teams already shipping AI features that need multi-provider access, failover, cost optimization, quota management or request visibility. It also suits teams that prefer self-hosting.
Good to know, and what to watch
- Supports four listed API protocols.
- Provides provider failover and ordered routing.
- API keys can have budget and rate limits.
- Self-hosted option uses SQLite by default.
- The free self-hosted plan requires deployment and management.
- Cloud usage is billed through top-up credits.
Plans
Every plan and priceSelf-hosted
Free plan
The verdict
YoloRouter brings protocol translation, routing controls and administration into one gateway, with self-hosted and cloud options. Teams should weigh its deployment model and usage-based cloud billing against their operational needs.
What the maker publishes
- Access controls
- API keys support model allowlists, rate and concurrency limits, cumulative budget caps, optional expiry, and instant revocation. source
- Access controls
- API keys can have model allowlists, rate and concurrency limits, budget caps, expiry and instant revocation. source
- API compatibility
- The API supports OpenAI Chat Completions, OpenAI Responses, Anthropic Messages, and Gemini generateContent protocols. source
- API reference
- The documentation provides an OpenAPI-generated reference with schemas, parameters and a live playground. source
- Architecture
- The backend uses Go with Gin, GORM and goose, while the Vue 3 and TypeScript frontend is embedded with go:embed. source
- Cloud billing
- YoloRouter Cloud uses top-up credits with no subscription and no minimum spend, charging users only for usage. source
- Cloud billing
- YoloRouter Cloud uses pay-as-you-go credits without a subscription or minimum spend. source
- Company identity
- The public GitHub organization identifies itself only as YoloRouter and does not state a legal name, headquarters or founding year. source
- Cost optimization
- YoloRouter can compress bulky tool output before forwarding it to upstream models and reports measured savings. source
- Cost optimization
- YoloRouter can inject system prompts and compress bulky tool output while reporting measured savings. source
- Deployment
- The product is distributed as a single binary with an embedded admin console and no Node runtime requirement. source
- External licence record
- The external mirror identifies the repository licence as Apache-2.0. source
- Install platforms
- Install documentation covers Docker, release binaries and service installers for Linux, macOS and Windows. source
- Key rotation
- Provider key pools automatically skip rate-limited, unauthorized or quota-exhausted keys. source
- Limit: input lengths
- MiniMax accepts up to 10,000 characters and Zhipu up to 1,024, with upstream errors returned for excess input. source
- Limit: media URLs
- Returned image and video URLs retain the upstream provider's expiry because YoloRouter does not rehost them. source
- Limit: Responses images
- Responses input_image entries are dropped when translation to another egress protocol is required. source
- Limit: speech failover
- Speech requests do not fail over because voice selections are vendor-specific. source
- Limit: video cancellation
- Video jobs have no cancellation endpoint in the supported task dialects. source
- Media APIs
- The gateway exposes OpenAI Images and Videos APIs beside chat and bills delivered images or delivered seconds. source
- Model catalogue
- The models page lists 21 available models with pricing, context windows and capabilities. source
- Model pricing
- The pricing page lists Claude Opus 4.8 at input $5→$1.5 and output $25→$7.5. source
- Notable news
- Release v0.2.8 on October 11, 2026 added session observability and replay, capability enforcement, admin audit and media playground APIs. source
- Observability
- The console reports token and cost KPIs, usage breakdowns and full per-attempt request logs with CSV export. source
Checked against the maker's pages on 15 Oct 2026.