What OpenRouter charges
OpenRouter is a hosted router with an OpenAI compatible API. A fee when you buy credits: 5.5% on Standard and 8% on Business. Token prices are the provider's list price. A Free plan with 25+ free models and 50 requests a day.
With your own keys (BYOK): Your provider bills you directly. The first $25,000 of list-price inference a month through your own keys has no OpenRouter fee; after that it is 5% of what the request would cost on OpenRouter.
As of . Check OpenRouter's pricing page for the current terms.
OpenRouter, FUSR.ai, and self hosted LiteLLM
The things that change your bill and your workload, side by side.
| OpenRouter | FUSR.ai | Self hosted LiteLLM | |
|---|---|---|---|
| Whose keys | OpenRouter credits, or your own keys | Your own Anthropic, OpenAI, and xAI API keys, stored encrypted in the proxy. | Your own keys |
| What it adds to token prices | A fee when you buy credits: 5.5% on Standard and 8% on Business. Token prices are the provider's list price. With your own keys, no OpenRouter fee on the first $25,000 of list-price inference a month, then 5% of what the request would cost on OpenRouter. | Nothing on tokens. Your providers bill you directly. A flat plan: Free $0, Pro $29 a month, Team $149 a month. | Nothing. It is open source; you pay for your server and database. |
| Prompt caching | Prompt caching is on Standard and up, not Free. | On every plan. A chat stays on the model whose cache is warm. | Not covered here. See the LiteLLM docs. |
| Spend caps | Budgets and spend controls are on Standard and up, and each API key can carry its own credit cap. | One monthly cap on Free. Daily and monthly caps on every key on Pro and Team. | Virtual keys with max_budget and budget_duration, team budgets, and rate limits. Budgets need a Postgres database. |
| Who runs it | Hosted for you | Hosted for you, on LiteLLM. | You do |
| API | An OpenAI compatible API. | OpenAI compatible, at https://proxy.fusr.ai/v1 with the model fusr/auto. | The OpenAI API |
As of . Check OpenRouter's pricing page for the current terms.
When OpenRouter is the better fit
- You need models from providers FUSR.ai does not route. FUSR.ai covers Anthropic, OpenAI, and xAI only.
- You want to try free models before you pay for any. OpenRouter's Free plan has 25+ of them, at 50 requests a day.
- You would rather buy credits in one place than hold a key with each provider.
When self hosted LiteLLM fits better
LiteLLM is free, open source, and good software. Virtual keys with max_budget and budget_duration, team budgets, and rate limits. Budgets need a Postgres database. Self host it if you're happy to run it: a server, a Postgres database, config, and upgrades. FUSR.ai runs on LiteLLM too. Hosted LiteLLM, explained.
When FUSR.ai fits
- You already pay Anthropic, OpenAI, or xAI directly and want to keep those keys and those bills.
- You want a flat plan instead of a fee that grows with your spend.
- You want each request sent to a model that fits with fusr/auto, prompt caching kept on, a cap on every key, and every pick logged with a reason.
- You don't want to run a proxy yourself.
FUSR.ai does not use chat plans like Claude Pro or ChatGPT Plus, and it routes only to Anthropic, OpenAI, and xAI. Plans and prices are on the pricing page.
Moving from OpenRouter
Both speak the OpenAI API, so moving is a base URL, a key, and a model name. Connect your provider keys on FUSR.ai, run the setup command, and ask for fusr/auto, or fusr/claude, fusr/openai, or fusr/grok for one provider every time. Keep OpenRouter until you are happy.
npx fusrai setupSetting a tool up by hand is in the setup docs.