How to use the Ntense AI API in Cloud Workspace and your own apps

Want to try an AI feature while learning or building in Ntense? The Ntense AI API is ready to use in a Ntense Cloud Workspace. If you are working outside the Workspace, create an API key in your Ntense account and use the same endpoints from your own app or terminal.

The API is provided as an out-of-the-box solution for study and experimentation, with limited usage. Choose a model for the task, make a small request, then inspect the result before building on it.

Which connection settings should I use?

• NTENSE_BASE_URL — OpenAI-compatible API base URL — https://ntense.ai/ai/v1
• NTENSE_MODEL — Default model for an OpenAI-compatible client — ntense-expert for text examples
• NTENSE_CLAUDE_BASE_URL — Anthropic Messages-compatible API base URL — https://ntense.ai/ai
• NTENSE_CLAUDE_MODEL — Default model for an Anthropic-compatible client — ntense-expert for text examples
• NTENSE_API_KEY — Authorises your API requests — Your own key from your Ntense account

The two base URLs serve different client formats. Point an OpenAI-compatible client at NTENSE_BASE_URL. Point an Anthropic Messages-compatible client at NTENSE_CLAUDE_BASE_URL. Use the matching model variable for the client you are configuring.

Using a Ntense Cloud Workspace

Open a terminal in your Workspace and use the existing NTENSE_BASE_URL, NTENSE_MODEL, NTENSE_CLAUDE_BASE_URL, NTENSE_CLAUDE_MODEL, and NTENSE_API_KEY environment variables directly. The Workspace provides its own API key, so you do not need to create a separate personal key for requests made there.

Your application can read these values from its environment. Keep the key on the server side; do not paste it into browser code, commit it to a repository, or print it in logs.

Using the API outside a Ntense Cloud Workspace

Sign in to Ntense, open Account → API Keys, and select Create API key. Copy the new key when it appears; the secret is shown only once. Configure the five environment variables above in your local environment or your app's secret store. Set NTENSE_API_KEY to the key you created.

If you no longer need a key, remove it from the same account page. Use a separate user-created key for work outside your Workspace rather than copying the Workspace key out of its environment.

This also works in a Google Colab or Kaggle notebook when internet access is enabled: keep your personal key in the notebook provider's secret storage and call the same API base URL from notebook code. You only need Ntense Tunnel when you want Ntense to work with tools in that environment, and only where the provider permits it. Google Colab FAQ^[1], Kaggle Notebooks documentation^[2].

Which model should I choose?

These are example model IDs, with one model shown for each task type:

• Text chat and reasoning — ntense-expert
• Image generation — ntense-image
• Audio conversation — ntense-audio
• Text to speech — ntense-tts
• Speech transcription — ntense-transcribe
• Text embeddings — ntense-text-embedding

Start with ntense-expert for the text request below. The other task types use their corresponding API operations; changing only the model name in a chat request is not enough to generate an image, transcribe audio, or create embeddings.

Make your first text request

Once the environment variables are available, run this command in your terminal:

curl "$NTENSE_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $NTENSE_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{\"model\":\"$NTENSE_MODEL\",\"messages\":[{\"role\":\"user\",\"content\":\"Explain one practical way to test an AI-generated answer.\"}]}"

The response contains the model's answer in the choices array. Check the answer against your task before you use it in a project. If the request is rejected, confirm that your key is active, the base URL matches the client format, and the model ID is set.

What is this API intended for?

This ready-to-use API supports study and experimentation with limited usage. It is a starting point for learning how to build and evaluate an AI-powered feature, while you remain responsible for testing the result and protecting your key.

We plan to offer a pay-as-you-go option for people who need more capacity, including video models in the future. Video models are not part of the current model list above.

If you would rather describe an idea and try a working app first, follow the Ntense app tutorial. To work with your local browser, development tools, or a permitted remote environment, read how Ntense Tunnel connects your own tools.

Sources and further reading

• Google Colab FAQ^[1]
• Kaggle Notebooks Documentation^[2]