Question 1

Do I have to rewrite my application?

Accepted Answer

No. go2llm is protocol-compatible with Anthropic Messages and OpenAI Chat Completions, so what changes in your application is the base URL and the key. Business logic, response handling and streaming stay exactly as they are.

Question 2

Which base URL do I use?

Accepted Answer

For the Anthropic SDK and Claude Code — https://go2llm.tech with no suffix. For the OpenAI SDK and OpenAI-compatible clients — https://go2llm.tech/v1.

Question 3

Can one key be used across different models?

Accepted Answer

Yes. A gk_sk_… key works with every published model and both protocols, and the balance is shared. If you need to, restrict the list of allowed models in the key's settings.

Question 4

Do I have to top up before the first request?

Accepted Answer

No. Signing up credits $1 to your balance, which covers the first few requests. Topping up makes sense once you have confirmed the integration works.

Question 5

Where do I see what a request cost?

Accepted Answer

The cost of every request comes back in the usage.cost_usd field of the response, including in the final usage block of a streamed response. A summary across all keys is in the dashboard.