Upload one character reference and place the same character in new scenes. Try Ideogram 3.0 Character on HiAPI with examples, three per-image tiers and async API samples.
Provider: Ideogram
Category: image generation
Endpoint: /v1/tasks
Status: Available
Cost: See live page pricing
Upload one character reference and place the same character in new scenes. Try Ideogram 3.0 Character on HiAPI with examples, three per-image tiers and async API samples.
Pricing
Standard Usage
What to exclude; Up to 5000 characters
TURBO is fastest, BALANCED is the default, QUALITY has the finest detail; each tier is priced separately
AUTO, REALISTIC or FICTION (illustrated)
Expand the prompt with MagicPrompt
Ready for generation
Configure your parameters and click "Run" to see the output here.
Each case separates references, the edit instruction, and the complete result, with image zoom and prompt copy.
1 reference image / 1328 × 752
The same explorer walks a lantern-lit market with hot noodles; a landscape reference yields a landscape frame.
The explorer character walking through a lantern-lit night market, holding a paper cup of hot noodles, warm glowing stalls on both sides, storybook illustration, no text, no signs
1 reference image / 1328 × 752
Sea wind, gulls and a red kite, with the yellow raincoat, round glasses and backpack kept consistent.
The explorer character flying a red kite on a wide windy beach, kite string stretching across the sky, seagulls, soft afternoon light, storybook illustration, no text
1 reference image / 1328 × 752
A photo reference with REALISTIC style places the same person into a café work scene.
The same man working as a barista in a sunlit neighborhood cafe, pouring latte art into a ceramic cup, wearing a canvas apron, shallow depth of field, photorealistic, no text
1 reference image / 864 × 1152
A rainy platform with wet reflections and warm lamps; a portrait reference produces an 864×1152 portrait output.
The explorer character reading a paper map on a rainy night train platform, glowing lamps, reflections on wet tiles, cinematic, no text, no signs
1 reference image / 864 × 1152
A bright summer field where the hair, outfit and expression style stay consistent.
The explorer character laughing in a vast sunflower field under a bright blue summer sky, holding the backpack straps, no text
1 reference image / 864 × 1152
BALANCED tier: a seaside ride where the character stays consistent in a dynamic pose.
The same explorer character riding a red bicycle along a seaside cliff road at sunset, wind in the hair, warm golden light, wide shot, no text
Choose an adjacent model by task, output tier, and reference-image needs.
Masked character edit
Inpaint the same character into an existing image while keeping the rest unchanged.
Typography and layouts
Choose Ideogram V4 for poster lettering, logos and signage.
Multi-reference generation and editing
Compare Nano Banana 2.1 when you need several references or free aspect ratios.
Upload one character reference and place the same character in new scenes. Try Ideogram 3.0 Character on HiAPI with examples, three per-image tiers and async API samples.
The Playground and API use the same model ID: ideogram-v3-character. Reference images provide subject, product, scene, or style facts, while the prompt assigns each role and limits the edit.
Choose inputs and output settings for your creative goal with Ideogram 3.0 Character.
Each request returns one image; references, source images and masks are not charged separately. Failed tasks are not charged; see live pricing.
References provide visual facts; the prompt assigns their roles and limits the edit.
Pick a clear image on a simple background; illustrations, mascots and photos all work. Its orientation sets the output frame.
Refer to the character with a fixed phrase such as the same boy, then add scene, action, props and light.
Explore
TURBO
Final
QUALITY
TURBO for exploring, BALANCED for everyday output, QUALITY for final art; style can be AUTO, REALISTIC or FICTION.
Copy a minimal request below. See Docs for parameters, callbacks, task retrieval, and errors.
POST /v1/tasksideogram-v3-charactercurl -X POST "https://api.hiapi.ai/v1/tasks" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "ideogram-v3-character",
"input": {
"prompt": "The explorer character walking through a lantern-lit night market, holding a paper cup of hot noodles, warm glowing stalls on both sides, storybook illustration, no text, no signs",
"reference_image_urls": [
"https://static.hiapi.ai/gallery/2026/10/156346d531f0b319.jpg"
],
"rendering_speed": "TURBO"
}
}'The create request immediately returns a taskId; image generation continues asynchronously.
ONE REFERENCE, ONE CONSISTENT CHARACTER
Ideogram 3.0 Character focuses on character consistency. Upload one character reference, describe a new scene, action and mood, and the model keeps the hairstyle, outfit, colors and facial features as it places the same character in a market, a beach, a platform or a café.
On HiAPI it comes as two models: ideogram-v3-character generates new images from the reference, and ideogram-v3-character/edit uses a source image and mask to repaint the character into a chosen area only. Enlarge any example on this page and copy its original prompt.
EXAMPLES AT A GLANCE
Swipe to browse. Each card shows the output ratio and lets you copy the full prompt.
16:9The explorer character walking through a lantern-lit night market, holding a paper cup of hot noodles, warm glowing stalls on both sides, storybook illustration, no text, no signs
16:9The explorer character flying a red kite on a wide windy beach, kite string stretching across the sky, seagulls, soft afternoon light, storybook illustration, no text
16:9The same man working as a barista in a sunlit neighborhood cafe, pouring latte art into a ceramic cup, wearing a canvas apron, shallow depth of field, photorealistic, no text
3:4The explorer character reading a paper map on a rainy night train platform, glowing lamps, reflections on wet tiles, cinematic, no text, no signs
3:4The explorer character laughing in a vast sunflower field under a bright blue summer sky, holding the backpack straps, no text
3:4The same explorer character riding a red bicycle along a seaside cliff road at sunset, wind in the hair, warm golden light, wide shot, no text
ONE REFERENCE, MANY SCENES
The single character reference is on the left. The three results only change the scene and action in the prompt: hair, freckles, earrings and sweater stay consistent across light and pose, while background and composition are entirely new.
RECURRING CHARACTER
Using one watercolor character reference, four prompts tell a story: find the map, cross the bridge, camp by the fire, reach the lighthouse. The red scarf, green jacket and satchel carry through every frame, ready for picture-book pitches, storyboards and series art.
01Finding a map in the attic
02Crossing a misty rope bridge
03Campfire under the stars
04Reaching the lighthouse at dawn
01 / CONSISTENCY
The reference shows an illustrated character in a yellow raincoat, round glasses and a canvas backpack. In the night market, the raincoat, glasses, backpack and teal hair remain; only the action, props and lighting change.
Refer to the subject with a fixed phrase such as “the explorer character” and spend the prompt on scene, action and light; this is steadier than re-describing the appearance.
02 / ASPECT RATIO
The output aspect ratio roughly follows the character reference: a 768×1024 portrait reference returns 864×1152, while a landscape reference returns 1328×752. There is no separate size parameter.
For a landscape cover or 16:9 frame, extend the reference to landscape first (for example by padding the background on both sides), then generate.
03 / PHOTO REFERENCE
With a photographic reference and style set to REALISTIC, the same person appears in a café pouring latte art, with the face, hair and shirt texture carried over.
For realistic scenes, specify lens, depth of field and light; the character should be clearly visible in the reference.
PARAMETER COMPARISON
Mascot reference, prompt and seed are identical; only rendering_speed changes. TURBO suits fast composition exploration, BALANCED is the default, and QUALITY adds finer fur, bunting and background detail. See live pricing for costs.
| Tier | Result | Best for |
|---|---|---|
| TURBO | Exploring directions, bulk drafts | |
| BALANCED | Everyday output (default) | |
| QUALITY | Final art, key visuals |
Same illustrated character, prompt and seed. The difference is mostly light and texture: REALISTIC leans toward photographic lighting and FICTION keeps a stronger illustrated look, while outfit and hair stay the same.
| Style | Result | Best for |
|---|---|---|
| AUTO | Let the model infer from the prompt | |
| REALISTIC | Portraits, products, realistic scenes | |
| FICTION | Picture books, games, cartoon characters |
Keep one hero across a whole book or storyboard while scenes change.
Use one mascot across billboards, packaging, seasonal posters and social posts.
Produce ongoing series of a virtual persona in new places and activities.
When one character must appear across many images, Ideogram 3.0 Character reduces per-image retouching.
Generate chapter scenes from one character sheet while keeping the lead's design.
Example: field, platform, market and beach scenes
Place a mascot into seasonal, store or event visuals without redrawing.
Tip: explore in TURBO, finalize in QUALITY
Generate work or lifestyle scenes from one portrait for storyboards and pitches.
Example: bookshop → café
These notes correspond to the examples on this page.
CHARACTER EDIT
Source + mask + character reference; everything outside the mask stays.
IDEOGRAM V4
Compare it for titles, logos and signage.
Upload a reference in the Playground above to generate. For integration, submit through the HiAPI /v1/tasks endpoint, billed per image with no charge for failures.
Not separately. The output follows the character reference: portrait references return 864×1152 and landscape references return 1328×752. Upload a landscape reference (for example by padding the background) for landscape output.
A clear frontal or three-quarter view on a simple background; illustrations and photos both work. Outfits and colors in the reference are kept where possible.
Refer to the subject with a fixed phrase (the explorer character / the same man) and focus on scene, action, props and light; pair it with style REALISTIC or FICTION.
Yes. The orange fox mascot on this page comes from one character design and appears on a billboard, behind a counter and at a fair with the same scarf and colors.
Yes. Use the same character reference for every frame and change only the scene and action in the prompt, as in the four-frame story on this page.
AUTO lets the model decide from the prompt; REALISTIC leans toward photographic light and texture for portraits and real scenes; FICTION keeps an illustrated look for picture books, games and cartoon characters.
A fixed seed makes results reproducible and comparable: keep the reference, prompt and seed, then switch only the tier or style to compare.
output[].url is a temporary link with an expiry; download or copy results to your own storage after the task succeeds.
TURBO is fastest and lowest-priced for exploring directions and drafts; BALANCED is the default; QUALITY gives the finest detail for final images. All three use the same model ID and are priced per image; see live pricing.
No. reference_image_urls must contain exactly one image. A clear full- or half-body image on a clean background gives the most stable results.
Not yet. Character generation (ideogram-v3-character) and masked editing (ideogram-v3-character/edit) are available today.
Failed tasks are not charged. Save the taskId, query GET /v1/tasks/:id and download the WebP from output[].url when status is success. In production, pass callback.url for terminal notifications and save results promptly.