Skip to main content
A tool that lets you set a base URL, a key and a model name can use NagaAI. The three settings below cover most coding agents, chat apps and SDKs.

​
Before you start

You need a NagaAI API key. If the tool has its own integration page, follow that page instead.

​
Connect the tool

1

Pick the base URL

Match the base URL to the format the tool speaks, and use your NagaAI API key as the key. If the tool offers a choice, pick Responses or Anthropic over Chat Completions.
2

Pick a model

Use a model id from naga.ac/models. Agents need a model whose chat operation lists tools in parameters. A tool that reads /v1/models lists every model, including image, audio and embedding models, so choose a chat model by name.
3

Check the connection

Send a short message. The request appears in Logs in your dashboard.
NagaAI drops unknown fields instead of failing, while functions and unknown tool types fail with 400. Differences from OpenAI and Anthropic lists what else changes.

​
Cursor

Cursor accepts an OpenAI key and a custom base URL under Settings → Models, with Override OpenAI Base URL set to https://api.naga.ac/v1. Per Cursor’s documentation, custom keys work only for chat models, and Tab completion keeps using Cursor’s own models. Cursor staff have confirmed on the forum that the override skips Claude and Gemini model names. For GPT model names Cursor sends a request format that NagaAI rejects. Add a model id from another publisher, such as minimax/minimax-m3.1-flash, and use that.

​
Gemini CLI

Gemini CLI has no official way to use an OpenAI-compatible or Anthropic-compatible endpoint. It talks only to Google’s own API format, which NagaAI doesn’t serve.

​
Troubleshooting

​
401 invalid_api_key

The tool sent no key or a wrong one. Paste your NagaAI API key into its key field. A provisioning key doesn’t work for model requests.

​
400 model_not_found

The tool sent a model id that NagaAI doesn’t serve. Copy the id from naga.ac/models.