xAI

xAI serves the Grok model family through its own API platform. Dline appears under Provider as xAI.

Website: x.ai | API console: console.x.ai

Get an API key

1. Go to console.x.ai and sign up or sign in.

2. Add billing details or credits to the team that will pay for requests.

3. Open the API keys page and create a new key.

4. Copy the key immediately and store it securely.

Configure Dline

See Model configuration for the general flow.

1. Click the Settings button in the Dline panel title bar.

2. Set Provider to xAI.

3. Paste the key into the xAI API Key field.

4. Select a model under Model.

Dline sends requests to https://api.x.ai/v1.

Models

Model	Context window	Notes
grok-4	262,144	Default model. Reasons before it answers, and its reasoning is not shown.
grok-4-1-fast-reasoning	2,000,000	Reasons before it answers, and its reasoning is not shown.
grok-4-1-fast-non-reasoning	2,000,000	Answers directly.
grok-4-fast-reasoning	2,000,000	Reasons before it answers, and its reasoning is not shown.
grok-code-fast-1	256,000	Coding-oriented.

The card also lists the older Grok 3 and Grok 2 families, plus grok-beta and grok-vision-beta. Pick one of those only when a task needs a model that the newer families no longer serve.

Reasoning

The xAI card has no thinking control, because the reasoning level belongs to the model ID: a reasoning model always reasons, and a non-reasoning model never does. Choose the variant you need under Model.

Note

Dline hides reasoning content from the Grok 4 family. Those models stream a thinking trace that restates the answer rather than explaining it, so showing it would only add noise to the chat.

Prompt caching

Models that declare supportsPromptCache accept a cached prefix. Dline reports the cached share of the input separately in the usage panel, so a repeated system prompt and workspace context are billed at the cached rate instead of the full input rate.

Troubleshooting

Problem	What to do
401 on every request	The key is not valid for https://api.x.ai/v1. Create a new key in the console and paste it again.
403 naming the model or the team	The team is not entitled to that model, or has no credits. Check the model list and the billing state in the console.
404 on a model	The model ID does not exist for this account. Pick another model from the list instead of typing the ID.
429	The account hit its rate limit. Wait, or reduce how much context a task sends.
Reasoning never appears in the chat	Expected for the Grok 4 family. Their reasoning is hidden on purpose; the answer is unaffected.