LIMITED TIME OFFER
Starter
Try every model, one key
$20
/ seat / month
You get
$40 of model use
Up to $3 every 5 hours and $10 a week.
Includes:
- Every model through one API
- Routing and fallbacks
- Usage and activity per key
- Credit limits per key
LIMITED TIME OFFERPasses for the first 1,000 users
Every AI model. Get 2x usage of what you pay.
Pick a pass for each seat. A $20 pass gets you $40 of model use every month, on every model, at each provider's list rate. Past a seat's limits, it keeps going on credits.
Passes
Each seat gets twice its price in model use every month, on every model, at each provider's list rate.
LIMITED TIME OFFER
Growth
For teams shipping to production
$50
/ seat / month
You get
$100 of model use
Up to $5 every 5 hours and $25 a week.
Everything in Starter, plus:
- 2.5x Starter usage per seat
- Guardrails
- Bring your own keys
- Credit limits per person
LIMITED TIME OFFER
Pro
For heavy, always-on workloads
$200
/ seat / month
You get
$400 of model use
Up to $15 every 5 hours and $100 a week.
Everything in Growth, plus:
- 10x Starter usage per seat
- Highest rate limits
- Priority routing at peak times
Every pass includes every model through one API key. Limited time offer for the first 1,000 users.
How your pass works
- 1
Your pass pays first
A seat's requests draw its pass's included usage, at each model's list rates.
- 2
Hit a limit? Credits take over
Each pass sets how much a seat can use every 5 hours and every week. Past either limit, the seat keeps going on credits until the window resets, up to the credit limit you set.
- 3
It renews every month
Included usage renews each month. What a seat doesn't use doesn't roll over.
Seats and teams
One seat per person
Any pass, in any mix. Every member holds one, and its usage is theirs alone. A key can serve several people: each one draws their own seat first, then credits.
Organization and workspaces
Billing, credits and seats live on your organization, shared by every workspace. Each workspace has its own keys, usage, routing and guardrails. Invite people as Admin, Developer or Consumer, and each role sees only what it needs.
Credit limits
Cap what anyone spends past their seat, per key, per person or per guardrail.
Bring your own provider keys
Your key, the provider's bill
Add a key you hold with a provider such as OpenAI or Anthropic. Requests to that provider run on your key and are billed by the provider. Hopscotch charges nothing for them. If your key stops working, the request fails. Hopscotch never charges your credit for it.
More on bring your own key
Look up a model's price
Per million tokens, at each provider's list rate. A pass's included usage is drawn at these prices too. Search by name or ID.
| Model | Input | Output | Cached input |
|---|---|---|---|
| Claude Sonnet 5.5Anthropic · anthropic/claude-sonnet-5-5 | $2.00 | $10.00 | $0.20 |
| GPT-6.1 SolOpenAI · openai-main/gpt-6.1-sol | $2.00 | $10.00 | $0.10 |
| MiMo-V2.6-FlashXiaomi · deepinfra/MiMo-V2.6-Flash | $0.14 | $0.28 | $0.003 |
| DeepSeek V4.1 FlashDeepSeek · deepinfra/DeepSeek-V4.1-Flash | $0.20 | $0.60 | $0.006 |
| DeepSeek-V4.1-FlashDeepSeek · deepseek/deepseek-flash | $0.30 | $1.20 | $0.006 |
| GPT-6 AstraOpenAI · openai-main/gpt-6-astra | $10.00 | $50.00 | $1.00 |
| Gemini 3.8 FlashGoogle · google-aistudio/gemini-3.8-flash | $0.75 | $3.75 | $0.07 |
| Muse Spark 1.3Meta · meta/muse-spark-1.3 | $1.25 | $4.25 | $0.15 |
Included with every account
- ModelsOne OpenAI-compatible API and one key for every model in the catalog
- Spend controlsBudgets for every key, teammate and workspace, plus a per-minute cap
- FallbacksBackup models, tried in the order you set
- Request logsA request log with each call's model, provider, tokens, cost and speed, exportable to CSV
- Usage analyticsUsage by model, provider, key and teammate
- PlaygroundA playground that runs one prompt on up to three models side by side
- Zero Data RetentionZero Data Retention by default
Help on Discord or at support@hopscotchlabs.ai
How billing works
Credits
Top up by card. A request past a seat's included usage, or on an account without a plan, draws down your credits at its model's list rates for input, output and cached tokens.
Pay for what you use
You're charged only for the tokens each request uses.
Receipts
Every purchase gets a receipt on your billing statement.
No markup
You pay each provider's list price per token, with no markup and no added fees.
Limits
A request over one of your limits is refused before it reaches the provider, so it costs nothing.
Cut-off streams
If a stream stops partway, you pay only for the tokens it produced.
Low balance
Hopscotch emails you when your balance drops below $5, or below an amount you choose. Auto top-up can refill it.
Pricing questions
What is a Hopscotch Pass?
A monthly plan for one seat. Each seat gets twice its price in model use every month, on every model, at each provider's list rate.
How much model use does each pass include?
Starter is $20 a seat a month for $40 of model use. Growth is $50 for $100. Pro is $200 for $400.
Is this a limited time offer?
Yes. Passes are a limited time offer for the first 1,000 users.
What happens when a seat reaches its limits?
Every pass has a limit over 5 hours and another over a week. Past either, the seat keeps going on your workspace credits until the window resets, up to the credit limit you set for that person.
Does unused model use roll over?
No. Each seat's included usage renews every month.
Can I still pay as you go?
Yes. Top up credits from $5 without a pass, and add promotion codes at checkout. A pass sits alongside your credits.
Do you mark up token prices?
No. You pay each provider's list price per token, with no markup and no added fees. How this compares with OpenRouter
Can I cap what we spend?
Yes. Set a credit limit on each key, each person and the workspace, and cap spend per minute. Requests over a limit are refused.
Do I pay for requests that fail?
Requests your limits refuse cost nothing. If a stream is cut off partway, you pay for the tokens it produced.
Is it free to use my own provider key?
Yes. The provider bills you, and Hopscotch charges nothing for those requests.
Do you offer enterprise deals?
Yes. Talk to us at support@hopscotchlabs.ai or on the contact page. It's the same product, with the deal set up with our team.
Pick a pass and send your first request
You need a key, the base URL and a model name. The rest of your OpenAI SDK code stays the same.