Control AI costs and access top models through one API
Use 500+ models from Anthropic, OpenAI, Google, and more without a separate integration, account, or bill for each provider. Set spend limits for every key, teammate, and workspace.
lattice % export HOPSCOTCH_KEY=YOUR_HOPSCOTCH_API_KEY lattice % curl https://api.hopscotchlabs.ai/v1/chat/completions \ -H "Authorization: Bearer $HOPSCOTCH_KEY" \ -H "Content-Type: application/json" \ -d '{"model":"anthropic/claude-sonnet-5","messages":[{"role":"user","content":"ship it"}]}' lattice % ▍
Sonnet first. GPT, then Gemini, if it fails.
- [1]request
- [2]fallback
- [3]fallback
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.
- One key
- Models from Anthropic, OpenAI, Google, and more on one base URL, billed to one balance.
- Named models
- The model you name is the model that runs, and the Activity log shows which provider served it.
- Endpoints
- Chat completions and the Responses API, plus a models endpoint that lists every model you can call.
Set spend limits and see every request
Give each key a credit limit that resets daily, weekly, or monthly, set monthly limits for teammates and the workspace, and cap how fast the account can spend. A request that would cross a limit is refused before it reaches the provider.
The Activity log lists every request and exports to CSV, and Usage breaks spend down by model, provider, key, and teammate.
Compare models on your own prompts
The playground runs one prompt on up to three models side by side, billed through your key like any other request. Compare the answers and what each one cost before you change your code.
500+ models, each with its price and providers
Each model lists its context window and its price per million tokens. Where several providers serve the same open-weight model, the catalog shows each one and its price.
Models
- anthropic/claude-sonnet-52.00 in · 10.00 out / 1M
1.0M context · effort
- openai/gpt-5.64.00 in · 20.00 out / 1M
1.0M context · effort
- google/gemini-3.6-flash1.50 in · 7.50 out / 1M
1.0M context
- deepseek/deepseek-v4-flash0.44 in · 1.32 out / 1M
1.0M context · 4 upstreams · effort
- moonshot/kimi-k33.00 in · 15.00 out / 1M
1.0M context · 3 upstreams · effort
- qwen/qwen3.8-max2.00 in · 6.00 out / 1M
1.0M context · 3 upstreams · effort
- anthropic/claude-fable-510.00 in · 50.00 out / 1M
1.0M context · effort
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.
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. There are no plans or subscriptions.
Point your base URL at Hopscotch
Create an account, add credit, and issue a key to send your first request.