Skip to content

The LLM router that optimizes your AI spend

Call 500+ models at each provider’s list price, with no markup, through one API. Set a spend limit for every key and teammate, with zero data retention built in.

Backed by
  • Hustle Fund
  • SBI
  • Alliance
[YOUR APP]
Write three subject lines for the launch email.
[ANSWER]

1. It's finally here. 2. Built for how you work. 3. Your first look, inside.

Refused before any provider saw it. Not billed.
gpt-6.1-sol via openai · rerouted once · 1.4s · $0.004Send
↑ ANSWER
HOPSCOTCH[ANSWERED]
maya · …a91c$4.20 / $10
[PROVIDER SPEEDS · LIVE]
deepinfra429 · down
openai974 ms
google1,424 ms
Anthropicdeepinfra[429]
OpenAIopenai[200]
Googlegoogle[STANDBY]
  1. No markup or fees

    You pay each provider's list price per token. See the full OpenRouter comparison

    See pricing
  1. Switch models by changing one line

    Set a new base URL and key in the OpenAI SDK you already use. After that, moving to another model means changing the model name, with no new SDK to install.

    Read the docs

What comes with every key

  1. Cap what every key can spend

    Give each key, teammate and workspace its own budget. When an agent loops, it hits the limit and stops. Refused requests cost nothing.

    More on spend controls
  2. Stay up when a provider goes down

    List your models in the order you want them tried. If the first one can't answer, the next one does.

    More on fallbacks
  3. Trace any request by its ID

    Every response and error carries a request ID. Look it up to see which provider served it, what it cost and how it ended.

    More on request logs
  4. Every request measured

    Usage shows what you spend, how fast requests come back and how often they fail. Use it to decide which models and providers to keep.

    More on usage analytics
  5. Get more from the keys you have

    Use the rates and credit you already have with OpenAI, Anthropic and other providers. Add limits, fallbacks and usage tracking. Hopscotch charges nothing on top.

    More on bring your own key
  6. Test your prompts

    Run one prompt on up to three models side by side. Compare the answers, speed and cost before you pick one.

    More on the playground

Why engineering leads move their team

You own the AI bill, the uptime and the security review. Hopscotch gives you one place to manage all three while your engineers keep shipping with the models they want.

  1. One account instead of one per provider

    Without a router, every engineer who wants to try a model signs up with another provider, and the company card ends up on a handful of dashboards. With Hopscotch the team draws on one balance, and every request, from every provider, shows up in one Usage view.

  2. Onboard in a minute, offboard in one click

    A new engineer gets a key with a daily limit on their first morning. When someone leaves, revoke their key and nothing else changes. No shared secrets to rotate across every provider console.

  3. No rewrite to adopt it, none to leave

    Hopscotch speaks the OpenAI SDK your code already uses. Adopting it is a new base URL and key. Leaving is changing them back, so trying it costs an afternoon, not a migration.

  4. Your prompts stay yours

    Zero data retention is on by default for every account, and Hopscotch never stores prompts or responses unless you opt in. Request logs keep the model, tokens, cost and timing, not the content.

  1. Streaming starts sooner on every model we measured.

    A public watcher sends the same prompt to the same five models through Hopscotch and Cheaper Inference every 5 minutes, and times the first streamed token.

    faster to the first token, at the median
    2.2x

Common questions

Do I have to change my code?

Just three settings: the base URL, your API key, and the model name, which starts with the provider, such as anthropic/claude-sonnet-5. The rest of your OpenAI SDK code stays the same.

Can I switch between models?

Yes. Change the model name in your request, for example from openai/gpt-5.6 to anthropic/claude-sonnet-5. You can move between any models in the catalog, from any provider, and your key, balance, and limits stay the same.

Will Hopscotch ever swap my model for a different one?

No. Hopscotch runs the model you ask for. If you want another model to take over when yours can't answer, list your backups in a routing profile, in the order you choose.

Do you change my prompts or the answers?

No. Your prompts reach the model as you wrote them, and the answers come back unchanged. Some features that providers run on their own servers, such as web search, audio, and hosted tools, aren't supported through Hopscotch's provider accounts. The docs list them all.

What happens if a provider has an outage?

Hopscotch retries your request first. If you use a routing profile, it then moves to the next model on your list. For chat requests, a final attempt runs your model through a backup provider. If every attempt fails, you get an error.

Do you mark up the token price?

No. You pay each provider's list price per token, with no markup and no added fees.

How does Hopscotch make money?

We buy inference from providers in bulk, at volume rates below their list price. You pay the list price, and we keep the margin between the two. That's how you pay list price with no markup or fees.

Do you charge for failed requests?

No. You won't be billed credits for failed requests.

Can I use my own provider keys?

Yes. Add your own key for a provider, such as OpenAI, and Hopscotch sends that provider's requests on your key. The provider bills you directly, and Hopscotch charges nothing for those requests. If your key fails, Hopscotch does not switch to its own key.

What stops an agent in a loop from draining the balance?

Spending limits. Set a limit for each API key, each teammate, and the whole workspace, and cap how much the account can spend per minute, $50 by default. Requests over a limit are refused, and an owner can pause all spending at once.

Can I see which model actually answered?

Yes. The Activity log shows the model and provider for every request, with its tokens, cost, speed, and result. Your code can also look up any request's tokens and cost through the API.

Do you store my prompts?

No. By default, Hopscotch never stores your prompts or the model's responses.

How do I pay?

Add prepaid credit by card, starting at $5. Auto top-up can refill your balance when it drops below an amount you choose.

Point your base URL at Hopscotch

Create an account, add credit, and issue a key to send your first request.

BASE URLhttps://api.hopscotchlabs.ai/v1