One API to access 500+ AI models from top providers with spend controls.
Hopscotch homepage hero
1 / 2Overview
What is Hopscotch?
500+ AI models available via a single API
Hopscotch is a unified API that gives developers access to 500+ AI models from Anthropic, OpenAI, Google, xAI, DeepSeek, Perplexity, and more — through one integration, one API key, and one bill. You pay each provider's list price per token with no markup and no subscription; credit starts at $5.
Swap one base URL and keep your existing OpenAI SDK and code. An OpenAI-compatible endpoint supports chat completions, the Responses API, and a models endpoint that lists every callable model with its context window and price. Routing profiles add automatic fallbacks, and the Activity log shows which model and provider served each request.
Keep costs under control with spend limits for every key, teammate, and workspace (caps can reset daily, weekly, or monthly, and over-limit requests are refused before reaching the provider). Usage analytics break spend down by model, provider, key, and teammate, and a model lab lets you compare models on your own prompts before changing code.
Pricing: pay-as-you-go only — buy prepaid credit starting at $5, pay provider list prices, no markup, no fees, no subscription. Bring your own provider keys free with no monthly cap (the provider bills you directly). No free trial seen on site.
Platforms and languages
Hopscotch Availability
Platforms
Languages
Capabilities
Hopscotch Key Features
Unified model API
One base URL, one API key, and one balance for 500+ models from Anthropic, OpenAI, Google, xAI, DeepSeek, Perplexity, and more.
OpenAI-compatible endpoint
Keep your existing OpenAI SDK and code — just change the base URL and model name (e.g. anthropic/claude-sonnet-5).
Named models
The model you name is the model that runs; the Activity log shows which provider served each request, with tokens, cost, speed, and result.
Routing and fallbacks
List backup models in a routing profile so requests fail over automatically when your primary model is unavailable.
Spend limits
Set credit limits per key, teammate, and workspace that reset daily, weekly, or monthly; over-limit requests are refused before reaching the provider.
Usage analytics
Break down spend by model, provider, key, and teammate, with request logs that export to CSV.
Model comparison lab
Run one prompt on up to three models side by side and compare answers and costs before changing your code.
Best for
Who uses Hopscotch?
Developers and teams building AI products
Call many AI models through one integration with cost controls and fallbacks
Plans and access
Hopscotch Pricing
Paid
Pay-as-you-go: prepaid credit from $5, provider list prices, no markup or subscription. Bring your own keys.
Common questions
Hopscotch FAQs
Does Hopscotch substitute or modify the model I request?
The provider's models. Hopscotch states it runs the model you ask for, never swaps it for a different one, and does not change your prompts or the answers.
How does Hopscotch make money if there's no markup?
You pay each provider's list price per token, with no markup and no added fees. Credit starts at $5, and there are no plans or subscriptions.
Can I bring my own provider keys?
Yes. Add your own key for a provider like OpenAI; Hopscotch routes those requests on your key, the provider bills you directly, and Hopscotch charges nothing for them.
How do I stop an agent in a loop from draining my balance?
Set spending limits per API key, per teammate, and for the whole workspace (reset daily, weekly, or monthly), cap per-minute spend, and pause all spending at once. Requests over a limit are refused before reaching the provider.
Do I have to rewrite my code to switch?
Changing three things: the base URL (https://api.hopscotchlabs.ai/v1), your API key, and the model name (e.g. anthropic/claude-sonnet-5). The rest of your OpenAI SDK code stays the same.
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
