Ideogram: Ideogram 4.5
ideogram-4-5- Price
- From $0.008
- Modalities
- Image
About Ideogram 4.5
Ideogram 4.5 is Ideogram's image model for generating and editing pictures, built for iterative work in which untouched regions stay intact across repeated edits. It makes images from text at 1K and 2K presets and edits a source image using up to four extra reference images and an optional mask. It is the one to pick when a picture will go through many revisions.
Where it works well
- Precise edit copies pixels the change did not touch straight from the source, so repeated rounds do not degrade the image.
- A mask limits where changes land: black marks the area to edit and white the area to keep.
- Up to four reference images guide an edit, three when a mask is also supplied.
- The edited result returns at the source image's own width and height instead of a resized canvas.
- Edit instructions can be plain sentences or structured JSON.
When to choose another model
- Aspect ratios for editing must sit between 1:6 and 6:1, which rules out extreme banners.
- With a mask, the reference limit drops to three because the mask takes one slot.
- It is a still-image model; motion, 3D, and vector output need other models.
Getting started
Create API key
Create a key in Console, then use it with every model on the platform.
Send your first request
Copy the example for your language and run it against the endpoint.
Image editingText to imageTokenLab endpointPOST/v1/images/generationscurl -X POST "https://api.tokenlab.sh/v1/images/generations" \ -H "Authorization: Bearer sk-xxx" \ -H "Content-Type: application/json" \ -d '{ "operation": "image-to-image", "model": "ideogram-4-5", "prompt": "A minimalist product photo of matte black headphones on a soft blue background.", "image_url": "https://example.com/source.png" }'
Parameters
Image editing
| Parameter | Values and limits | Default | Notes |
|---|---|---|---|
height | Not provided | Not provided | For source-image requests, supply width and height together as integers, each at least 256 and divisible by 32, with width*height <= 4194304 and aspect ratio from 1:6 to 6:1. Omitted dimensions have no shared fixed default: one delivery path selects a supported size automatically, while another uses 1024x1024. |
image | Not provided | Not provided | Multipart source-image file input. Repeat image or image[] for multiple files. Source inputs are combined into one ordered list of 1 to 5 images. The first image is edited; the remaining images guide the edit. Use JPEG, PNG, or WEBP, at most 25 MiB per image. Source aspect ratios must be from 1:6 to 6:1. |
image_url | Not provided | Not provided | One source-image URL. Source inputs are combined into one ordered list of 1 to 5 images. The first image is edited; the remaining images guide the edit. Use JPEG, PNG, or WEBP, at most 25 MiB per image. Source aspect ratios must be from 1:6 to 6:1. |
image_urls | Not provided | Not provided | A list of source-image URLs. Source inputs are combined into one ordered list of 1 to 5 images. The first image is edited; the remaining images guide the edit. Use JPEG, PNG, or WEBP, at most 25 MiB per image. Source aspect ratios must be from 1:6 to 6:1. |
images | Inputs: ≥ 1, ≤ 5 | Not provided | Limits apply when this input is supplied. Image aliases are combined into one input image list; the limit applies to the combined list. |
n | 1, 2, 3, 4 | 1 | Integer from 1 to 4; default 1. Counts 2 to 4 require PNG output and a delivery path that supports multiple results. A combination with no compatible delivery path fails before generation. |
operationRequired | Not provided | Not provided | Not provided |
output_format | png, jpeg, webp | Not provided | Use png, jpeg, or webp. JPEG and WebP require n=1. Omitted output_format has no shared fixed encoding: delivery paths return PNG or JPEG. Set it explicitly when the encoding matters. The spelling jpg is not accepted by the public endpoint. |
prompt | Not provided | Not provided | Required non-blank prompt or edit instruction, from 1 to 10000 characters. Prompts above 10000 characters are not supported by either delivery path. |
quality | very_low, low, medium, high | very_low | Not provided |
seed | ≥ 0, ≤ 2147483647 | Not provided | Optional integer from 0 to 2147483647. Omit it to let the model choose a random seed. No fixed seed default is asserted. |
width | Not provided | Not provided | For source-image requests, supply width and height together as integers, each at least 256 and divisible by 32, with width*height <= 4194304 and aspect ratio from 1:6 to 6:1. Omitted dimensions have no shared fixed default: one delivery path selects a supported size automatically, while another uses 1024x1024. |
Text to image
| Parameter | Values and limits | Default | Notes |
|---|---|---|---|
height | Not provided | Not provided | Without source images, supply width and height together as one of these exact pixel pairs: 2048x2048, 1440x2880, 2880x1440, 1664x2496, 2496x1664, 1792x2240, 2240x1792, 1440x2560, 2560x1440, 1600x2560, 2560x1600, 1728x2304, 2304x1728, 1296x3168, 3168x1296, 1152x2944, 2944x1152, 1248x3328, 3328x1248, 1280x3072, 3072x1280, 1024x3072, 3072x1024, 1024x1024, 896x1120, 1120x896, 864x1152, 1152x864, 832x1248, 1248x832, 800x1280, 1280x800, 720x1280, 1280x720, 720x1440, 1440x720. Omitted dimensions have no shared fixed default: one delivery path selects a supported size automatically, while another uses 1024x1024. |
images | Not provided | Not provided | Text-to-image does not accept source images. Do not send this field for that operation. Use one of this model's declared image operations for source-image requests. |
n | 1, 2, 3, 4 | 1 | Integer from 1 to 4; default 1. Counts 2 to 4 require PNG output and a delivery path that supports multiple results. A combination with no compatible delivery path fails before generation. |
operationRequired | Not provided | Not provided | Not provided |
output_format | png, jpeg, webp | Not provided | Use png, jpeg, or webp. JPEG and WebP require n=1 and quality=low or medium. High requires PNG output. Omitted output_format has no shared fixed encoding: delivery paths return PNG or JPEG. Set it explicitly when the encoding matters. The spelling jpg is not accepted by the public endpoint. |
prompt | Not provided | Not provided | Required non-blank prompt or edit instruction, from 1 to 10000 characters. Prompts above 10000 characters are not supported by either delivery path. |
quality | low, medium, high | low | Use low, medium, or high; default low. The gateway sends the selected tier explicitly. High requires PNG output. The text-to-image operation rejects source images; use operation=image-edit for source-image editing. |
seed | ≥ 0, ≤ 2147483647 | Not provided | Optional integer from 0 to 2147483647. Omit it to let the model choose a random seed. No fixed seed default is asserted. |
width | Not provided | Not provided | Without source images, supply width and height together as one of these exact pixel pairs: 2048x2048, 1440x2880, 2880x1440, 1664x2496, 2496x1664, 1792x2240, 2240x1792, 1440x2560, 2560x1440, 1600x2560, 2560x1600, 1728x2304, 2304x1728, 1296x3168, 3168x1296, 1152x2944, 2944x1152, 1248x3328, 3328x1248, 1280x3072, 3072x1280, 1024x3072, 3072x1024, 1024x1024, 896x1120, 1120x896, 864x1152, 1152x864, 832x1248, 1248x832, 800x1280, 1280x800, 720x1280, 1280x720, 720x1440, 1440x720. Omitted dimensions have no shared fixed default: one delivery path selects a supported size automatically, while another uses 1024x1024. |
Fields outside this contract are rejected.
quality selects the price tier: text-to-image low (default), medium, high; image-edit very_low (default), low, medium, high.
Pricing
The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.
Text to image
per image- Official price
- $0.03
- Verified price
- —
- Discount
- —
Text to image · Quality · Low
per image- Official price
- $0.03
- Verified price
- —
- Discount
- —
Text to image · Quality · medium
per image- Official price
- $0.06
- Verified price
- —
- Discount
- —
Text to image · Quality · High
per image- Official price
- $0.10
- Verified price
- —
- Discount
- —
Image editing
per image- Official price
- $0.008
- Verified price
- —
- Discount
- —
Image editing · Quality · very_low
per image- Official price
- $0.008
- Verified price
- —
- Discount
- —
Image editing · Quality · Low
per image- Official price
- $0.03
- Verified price
- —
- Discount
- —
Image editing · Quality · medium
per image- Official price
- $0.06
- Verified price
- —
- Discount
- —
Image editing · Quality · High
per image- Official price
- $0.22
- Verified price
- —
- Discount
- —
| Official priceper image | Verified priceper image | Discount | |
|---|---|---|---|
| Text to image | $0.03 | — | — |
| Text to image · Quality · Low | $0.03 | — | — |
| Text to image · Quality · medium | $0.06 | — | — |
| Text to image · Quality · High | $0.10 | — | — |
| Image editing | $0.008 | — | — |
| Image editing · Quality · very_low | $0.008 | — | — |
| Image editing · Quality · Low | $0.03 | — | — |
| Image editing · Quality · medium | $0.06 | — | — |
| Image editing · Quality · High | $0.22 | — | — |
Usage & activity
Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.
Usage & availability
Last 24 hours- Requests
- Success rate
- P95 latency
- Total tokens
Data is based on aggregate user requests, excluding status checks.
Open in Console
Open Ideogram 4.5 in Console with a prompt ready to edit or send.
Help me create an image with ideogram-4-5 using image-edit at /v1/images/edits. Show the result, cost, and any limits I should know.
Use cases
Best for- Text to Image
Product retouching
Recolour a bag, swap a label, or relight a packshot over several rounds while the rest of the photograph stays exactly as shot.
Campaign variations
Take one approved key visual and produce localized or seasonal versions by editing only the copy area or background.
Interior restyling
Mask a sofa or wall in a room photo and replace it with a reference item, keeping the rest of the room unchanged.
Prompt examples
Change the jacket in the photo to deep green corduroy, keep the face, pose, and background untouched.
A minimalist skincare poster, matte cream background, the product bottle centered, headline 'Slow Mornings' in thin serif type.
Using the attached skeleton reference, pose the dancer with her left arm raised and keep the studio backdrop.
FAQ
What is Ideogram 4.5 best at?
Iterative editing. Ideogram built it so that repeated changes keep untouched regions intact, which helps product retouching, relighting, recolouring, text swaps, and furniture replacement. It also generates new images from text.
Does Ideogram 4.5 support masks?
Yes. Supply a grayscale mask where black is the area to change and white is the area to keep. Masked requests can carry up to three reference images alongside the source.
What does precise edit do?
It leaves pixels unchanged where the edit did not meaningfully alter the image, copying them exactly from the original, and returns the output at the original width and height. That keeps multi-turn edits from drifting.
What output sizes does text-to-image offer?
The catalog describes 1K and 2K presets for generation from text. Edits keep the dimensions of the source image, so your input resolution decides the edited output.
How much does Ideogram 4.5 cost?
On TokenLab, Ideogram 4.5 costs $0.008-$0.22 per image. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.
Which endpoint should Ideogram 4.5 use?
Use https://api.tokenlab.sh/v1/images/edits for Ideogram 4.5. The request example below shows the matching code shape.
Which operations does Ideogram 4.5 support?
Ideogram 4.5 supports Image editing, Text to image. Select an operation above to see its endpoint and request example.
Compare Ideogram 4.5
Sources
Reviewed Oct 2, 2026