All use cases
Cost per token
Shop for the cheapest model without lock-in
Compare list prices in the catalog, then route through one URL so switching models is a string change.
What this is for
The cheapest SKU on a landing page is often the wrong unit (cached vs uncached, input vs output) or a model that is not actually serving.
Steps, then copy
- 1.Compare models on /models and /compare — use the published list prices, not screenshots.
- 2.Start traffic on anyrouter/auto so health-aware failover avoids a dead cheap SKU.
- 3.Pin a specific id only after you have measured quality on your prompts.
curl · anyrouter/auto
bash
curl https://anyrouter.dev/api/v1/chat/completions \
-H "Authorization: Bearer $ANYROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "anyrouter/auto",
"messages": [{ "role": "user", "content": "Say hello in one sentence." }]
}'Ready to try it
The guide has the full walkthrough. Sign up if you still need a key.
Next playbooks
- Start on anyrouter/free or anyrouter/autoTwo built-in virtual models: free failover on every plan (anyrouter/free) and stable paid failover (anyrouter/auto). Do not pick a flaky SKU for your first request.
- One API in front of every provider dialectOpenAI /chat/completions, Anthropic /messages, and a Responses API — same key, same catalog.
- Go: paid models from $2 a monthThe Go plan is $2/mo and adds paid-model access with a monthly credit. Free models have no daily limit on any plan. Cancel any time. Donating a provider key can unlock Go without paying.