Zils/ Docs

Appearance

Make your first API call ​

Call the shared JevK5 4B model without downloading weights or running a GPU. You need a Zils account with API access and a command line with Bash and curl.

1. Create an API key ​

Sign in at zils.ai/train, open API keys, and create a named key. Save the secret when it is shown; it is displayed only once. Keep it in your server's secret storage, away from browser code and source control. If your account lacks access, contact Zils.

Key management uses your signed-in session. Prediction calls use the resulting Zils API key. These credentials are not interchangeable.

In a Bash session, read the key without displaying it or including it in the command history:

bash
export ZILS_URL='https://training.zils.ai/decision'
read -r -s -p 'Zils API key: ' ZILS_API_KEY
export ZILS_API_KEY

https://training.zils.ai/decision is the hosted Zils decision API base. Training uses https://training.zils.ai.

2. List available models ​

bash
curl --fail-with-body "$ZILS_URL/v1/models" \
  -H "Authorization: Bearer $ZILS_API_KEY"

The shared alias is zils-shared. Available private models are scoped to your account. Use an immutable model ID from the catalog when you need to pin a release.

3. Ask a typed question ​

bash
curl --fail-with-body "$ZILS_URL/v1/systemone" \
  -H "Authorization: Bearer $ZILS_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"model":"zils-shared","state":"Please send the invoice for my last order.","questions":{"route":{"type":"choice","instructions":"Choose the responsible team.","criteria":{"billing":"Invoices and payments","technical":"Software troubleshooting","account":"Account access"}}}}'

The response's answers.route contains the selected choice, per-option probabilities, and confidence. The top-level model identifies the resolved release; usage records input tokens and zero generated output tokens. Read the actual returned values rather than assuming this example establishes accuracy on your traffic.

The API also supports yes/no (noul) and score questions. How decisions work explains the fields and their meaning.

Continue with your application ​

The API reference covers key creation/revocation, Python and JavaScript SDKs, bulk jobs, errors, capacity, and retention. Prepare a dataset covers task evaluation and optional training.

To self-host, use the API deployment guide. To run the upstream model directly, follow local JevK5 inference. The upstream runtime and Zils gateway have different request validation and confidence contracts.