Ideogram API pricing: per-image costs, credits and keys

Pay as you go vs monthly subscription plans

If you just want raw API access without a recurring seat commitment, the Pay as you go plan sells on-demand credits at $0.01 per credit. It includes API access with credit-based usage, but it limits you to 1 running or queued generation in the web app.

Subscription plans bundle monthly credits at discounted rates, though they are designed around web app seats:

• Ideogram Plus: $20 per month billed monthly, or $15 per month ($180 per year) billed annually, including 2,400 priority credits per month.
• Ideogram Pro: $60 per month billed monthly, or $42 per month ($504 per year) billed annually, including 7,500 priority credits per month.
• Ideogram Team: $30 per user per month billed monthly, or $20 per user per month ($240 per user per year) billed annually, including 3,600 priority credits per user per month. Team plans require at least 2 users.
• Legacy Basic: Ideogram no longer sells this plan ($8 monthly or $7 monthly billed yearly), but existing subscribers keep their grandfathered terms and get 960 credits each month.

For larger setups, Ideogram provides two Enterprise models. The Postpaid model requires a custom credit volume negotiated via sales. The Prepaid model is self-serve, offering custom upfront prepaid credits, pooled organization balances, and per-user spending caps.

All payments go through Stripe using credit or debit cards from Mastercard, VISA, or American Express. Taxes aren't included in listed prices, payments are non-refundable, and payment obligations are non-cancellable. If Ideogram makes material changes to API pricing, the terms require them to notify existing developers at least 14 days in advance via email or other agreed channels.

Is there a free trial for the API? No. Subscription plans offer no free trials, and accounts created with email and password get zero free credits. The only free tier is for app users who sign in via Google, Apple, or Microsoft, who receive 30 free credits per week.

What each image costs

Ideogram does not bill API calls as flat API requests. Instead, every operation consumes credits or bills as an itemized unit based on the model, speed, and endpoint you call.

You can check current per-operation costs inside the app by going to Settings → Usage costs, or by looking at the Pricing tab in the separate developer platform.

{
  "product": "ideogram_v4_generation",
  "endpoint": "/v2/image/generate/ideogram-4",
  "description": "Ideogram v4 Generation",
  "cost_total": "84.00",
  "currency_code": "USD",
  "source": "api",
  "dimensions": {
    "rendering_speed": "TURBO"
  },
  "billed_units": {
    "unit": "image",
    "quantity": "2100",
    "unit_price": "0.04"
  }
}

In the official usage endpoint documentation, an Ideogram v4 Generation request (ideogram_v4_generation) dispatched to /v2/image/generate/ideogram-4 with rendering_speed set to TURBO has a unit_price of $0.04 USD per image. In that example, rendering 2,100 images results in a cost_total of $84.00. Metered endpoints like gpt_image_2_edit do not report individual billed unit prices; they simply report the metered cost_total directly.

The default rate limit across the Ideogram API is 10 in-flight requests. If your backend tries to push 15 concurrent generation jobs, five will be rejected until the in-flight batch drops.

Workspaces, keys, and which balance gets billed

Creating an API key in Ideogram requires paying close attention to which workspace you are currently in. When you generate a key, you can enter an optional label up to 64 characters long. You get to see the secret key once. Copy it immediately; you cannot view it again later.

Where the invoice goes depends on the workspace where the key was generated:

• Personal workspace: The web app and API share the exact same credit pool. API calls spend your monthly subscription credits first. Once those run dry, calls spend your purchased top-up credits.
• Team workspace: An API key charges the individual credit balance of the specific team member who created that key. If an engineer creates a key under their seat, team jobs consume that engineer's monthly 3,600 credits, not a pooled company wallet.
• Enterprise Prepaid / API workspace: Keys burn directly against the organization's shared prepaid credit balance.
• Enterprise agreement: Keys bill against the negotiated company billing agreement.

Switching workspaces in the web UI does not redirect an existing key. Keys stay permanently bound to the workspace where you generated them. If you want a key to draw against a different workspace balance, you have to switch workspaces and generate a brand-new key there.

Revoking a key is immediate and permanent. It stops all future requests on that key right away. You cannot reactivate or undo a revoked key, though the revoked key ID remains selectable in the usage dashboard dropdown so you can still pull historical audit reports.

Do monthly credits roll over? No. Unused priority credits tied to monthly subscriptions expire at the end of each billing cycle. If you don't use them, you lose them. Purchased top-up credits, however, do roll over from month to month and only burn after your subscription credits are exhausted.

What happens when credits run out

Because Ideogram has no automated spending caps or native budget stop switches on developer accounts, understanding depletion rules keeps your production pipelines from breaking.

If you are on Pay as you go, running out of credits halts your generations. To resume service, you must manually buy more credits, or turn on auto recharge under Wallet & Billing so your card refills the balance when it empties.

If your API key is tied to a Plus, Pro, or Team workspace and runs out of monthly and top-up credits, generations using base Ideogram models drop into a slower queue that has no generation limit. Users on the legacy Basic plan also get slow queue access, but with a daily cap.

Slow queue fallback does not cover every API task. Even if you hold an active Plus, Pro, or Team subscription, the following features always require available credits and will not execute in the slow queue:

• Custom models
• Partner models
• Generations using a reference image
• Image edits
• Upscales

If your priority credits hit zero and you call an upscale or image edit endpoint, the call fails.

Reading spend in the UI and via the API

How you view your spend depends on your account tier. Personal accounts and seat-based Team plans do not have access to the in-app Usage page. The Usage screen in integrated Settings is strictly restricted to owners and admins of Enterprise and prepaid API workspaces.

The UI Usage dashboard defaults to the Last 30 days. It reports all timestamps in UTC and lets you switch between four timeframes: Month to date, Last 90 days (broken into daily buckets), Last 30 days (daily buckets), and Last 3 days (broken into hourly buckets).

To track spend programmatically, query the usage endpoint:

curl -X GET "https://api.ideogram.ai/v2/account/usage?start_time=2026-11-01T00:00:00Z&bucket_width=1d" \
  -H "Api-Key: YOUR_API_KEY"

The endpoint requires an API key owned by an organization admin. If a standard team member's key queries this endpoint, the API returns an HTTP 404 Not Found error.

Query rules and behavior for GET /v2/account/usage:

1. Required parameters: start_time formatted as an RFC 3339 UTC string.
2. Optional parameters: end_time, bucket_width (1d default, up to 92 days per call; or 1h, up to 168 hours), and sources (api, app, defaulting to both).
3. Data delays: Data lags live traffic by a few minutes, and responses are cached briefly, making the endpoint suitable for batch accounting rather than second-by-second traffic policing.
4. Returned fields: Line items return cost_total as a decimal string along with the ISO 4217 currency_code (USD). Products billed per item also supply billed_units containing the unit label, quantity, and unit price.

Tracking Ideogram spend with Costbrake

Because Ideogram lacks an API-driven spending limit and auto recharge can run unchecked on active workloads, tracking your burn rate is up to you.

Costbrake monitors spend across cloud, AI, and SaaS platforms like AWS, Google Cloud, Azure, Cloudflare, Vercel, OpenAI, Anthropic, and Ideogram. For Ideogram, Costbrake polls GET /v2/account/usage to read usage and spend per product and per API key in hourly buckets. Because that data can trail live traffic by a few minutes, every check re-reads the last 6 hours, and every 6 hours it re-reads the last 48.

To use it with Ideogram, you supply an API key from an organization owner or admin, tied to an Enterprise or prepaid API workspace (since seat-based Team plans don't expose usage endpoints). Ideogram reporting is USD only.

Ideogram does not have brakes inside Costbrake. Brakes are manual, reversible actions that only run when an admin clicks and confirms, and because Ideogram has no API for spending limits and revoking an API key is permanent, brakes are not available for Ideogram.

Instead, Costbrake catches billing surprises through alerting:

• Spike alerts: Triggers when an hour costs more than 3x your 7-day baseline and at least $5 more than normal.
• Budget thresholds: Alerts at 50%, 80%, and 100% of defined spend limits.
• Forecasts: Flags when month-to-date charges plus your recent 72-hour pace are projected to exceed your month-end budget.
• New charges: Alerts when an unfamiliar product line item passes $1.

Checks run hourly on the Free plan ($0 for 2 connections, 3 budgets, email alerts) and every 5 minutes on the Pro plan ($12/month or $120/year for 10 connections, 50 budgets, Slack, Discord, and signed webhooks). Team runs $39/month ($390/year) with PagerDuty. You can connect providers manually or point your AI coding agent to agent setup.

If you are comparing visual AI providers, see our guides on Stability AI API pricing: what a credit costs, what each image or audio call burns and how to watch your credit balance, How does fal.ai pricing work?, and How do I monitor spend across OpenAI, Anthropic, Gemini and other AI APIs in one place?.