Lumen
Lumen has been updated, please send email to help@ncsa.illinois.edu with subject Lumen if you have questions. Certain models will need to be acknowledged before use (one time only).

Profile & API Keys

The Profile page (/profile) shows your coin balance, spending history, model access, and your personal API keys.

Profile page

Profile Card

At the top of the page, a card shows your identity and a summary of your account activity.

The left side displays your avatar (pulled from Gravatar based on your email), your display name, and email address.

The right side shows six stat tiles:

Tile Description
Groups The groups you belong to
Coins Used Total coins spent across all models, all time
Tokens Used All input + output tokens across all models, all time
Favorite Model The model you have used most
Coins Available Your current pool balance with a progress bar (when a limit is set)
Refill Rate Auto-refill rate per hour and a countdown to the next refill

Coin Pool Values

Value Meaning
A positive number Coins remaining in your budget
Unlimited No cap — you can always send requests
No budget has been set up for your account
0 or negative Budget exhausted; further requests are blocked until a refill or an admin grants more coins

Web Chat Usage

Below the stat cards, a row summarizes your web chat activity:

Column Description
Conversations Number of conversations you have started
Requests Total messages sent through the chat interface
Tokens Total input + output tokens via web chat
Coins Total coins spent on web chat
Last Used When you most recently sent a message

API Keys

API keys let you access Lumen's AI models from your own code, scripts, or compatible tools — without opening a browser.

What an API Key Is

An API key is a secret token in the format sk_.... It identifies you to the API the same way your login session identifies you in the browser. Each key has its own usage counters, so you can track exactly how much each integration is using.

Important: The key is shown only once when you create it. Copy it immediately — it cannot be retrieved later.

Creating a Key

  1. Click + New API Key at the top of the API Keys section.
  2. A dialog opens and displays your new key in a read-only field.
  3. Copy the key now.
  4. Enter a descriptive name (e.g., my-research-script or jupyter-notebook).
  5. Click Save Key. The key now appears in the table.

Viewing Your Keys

The API Keys table shows all your active keys and lets you sort by name, requests, tokens, cost, or last used. Enable Show deleted keys to see previously revoked keys (displayed with strikethrough).

Column Description
Name The label you chose
Hint First 4 + last 4 characters of the key, for identification
Requests Total API calls made with this key
Tokens Total input + output tokens
Coins Total coins spent
Last Used Timestamp of the last API call
Actions Revoke button for active keys

Revoking a Key

Click Delete on any active key. The key is deactivated immediately — any code using it will start receiving authentication errors. Usage history is preserved and visible with "Show deleted keys".

Model Access

The Model Access table lists every model available in Lumen and your access status for each.

Column Description
Model Clickable link to the model detail page
Requests / Tokens / Coins / Last Used Your personal usage stats for that model
Access Your current access level (see below)
Status Health of the model's backend

Access Levels

Badge Meaning
Need Consent (warning) Model requires a one-time acknowledgment — click to enable it
Consented (green) You have acknowledged this model and can use it
Allowed (green) Model is fully available to you
Blocked (red) Model is not available to you

Model Status

Badge Meaning
ok (green) All backends healthy
degraded (yellow) Some backends are down but at least one is working
down (red) No healthy backends
disabled (gray) Model is currently turned off

Click any column header to sort the table. Use the search box to filter by model name. Check Show disabled to include currently disabled models.

National Center for Supercomputing Applications

Lumen

Illinois Computes GitHub Repository Request Feature