> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Ideogram V3

> Ideogram V3 is a text-to-image model optimized for rendering accurate typography inside generated visuals. It is intended for developers and designers building workflows that require precise text placement, logos, and packaging mockups.

## Overview

Ideogram V3 is a text-to-image model from Ideogram that specializes in generating legible, correctly spelled text within images. It accepts text prompts and outputs raster images, making it a focused tool for typographic design tasks.

On Kyma, the model runs through an OpenAI-compatible endpoint with a single API key. It supports prompt caching, which bills repeated prompt prefixes at this model's cached input rate, and includes automatic request failover. Exact costs are returned in the usage.cost field, and the X-Kyma-Model header reports the active model.

The model does not support image input, reasoning traces, or structured JSON outputs. It operates at a medium speed tier and is billed per image at a premium cost level.

## Specs

| Field              | Value                                 |
| ------------------ | ------------------------------------- |
| Model ID           | `ideogram-v3`                         |
| Best for           | Typography, packaging, posters, logos |
| Context window     | 0                                     |
| Input modalities   | Text                                  |
| Output modalities  | Image                                 |
| Tool calling       | Yes                                   |
| Structured outputs | Yes                                   |
| Prompt caching     | Yes                                   |
| Speed              | Medium                                |
| Cost band          | Premium                               |
| Release stage      | Stable                                |

## Pricing

|        | Per 1M tokens |
| ------ | :-----------: |
| Input  |     \$0.20    |
| Output |     \$0.60    |

## Use this when

* **Typography and Poster Design** — Generates posters and banners with correctly spelled, well-formatted text.
* **Product Packaging Mockups** — Creates packaging visuals with accurate brand names and label text.
* **Logo Concept Generation** — Produces logo drafts that integrate custom text cleanly into graphic elements.
* **Marketing Asset Creation** — Builds promotional images where headline text must render precisely.

## Not ideal for

Avoid this model for general-purpose photography, image-to-image editing, or workflows that require structured JSON outputs or vision input.

## Pick something else when

* You need photorealistic scene generation: use [`imagen-4-ultra`](/models/imagen-4-ultra).
* You require high-speed batch generation: use [`imagen-4-fast`](/models/imagen-4-fast).
* You need scalable vector design assets: use [`recraft-v4-vector`](/models/recraft-v4-vector).
* You want balanced quality for general art: use [`flux-2-pro`](/models/flux-2-pro).

## Example

```python theme={null}
from openai import OpenAI

client = OpenAI(base_url="https://kymaapi.com/v1", api_key="ky-...")

response = client.chat.completions.create(
    model="ideogram-v3",
    messages=[{"role": "user", "content": "..."}],
)
```

## Agent query example

Ask the API which models fit, instead of hardcoding an id:

```bash theme={null}
curl "https://kymaapi.com/v1/models?recommended_for=generate-images&tools=true&quality_tier=frontier-open"
```

## FAQ

**Does this model accept image inputs?**

No, it only accepts text prompts and outputs images.

**How are requests billed on Kyma?**

Requests are billed per image, with exact costs returned in the usage.cost field.

**Can I cache prompts to reduce costs?**

Yes, prompt caching is supported and bills repeated prefixes at this model's cached input rate.