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

# Grok Imagine 2 Image API Launch: Models, Pricing, and Integration

> LaoZhang API launched Grok Imagine 2 on August 13, 2026 with five aspect ratios, 1K/2K, up to 10 outputs, reference editing, and model-specific multi-reference limits.

* **Published**: August 13, 2026
* **Last verified**: August 13, 2026
* **Current status**: Active; current groups and prices are shown in the console

LaoZhang API now provides xAI's current Grok Imagine image generation through `grok-imagine-image` and `grok-imagine-image-quality`. Both models use the Images API and support five currently documented aspect ratios, 1K/2K output, up to 10 generated outputs per request, and reference-image editing. Current per-image prices are **\$0.02** and **\$0.045**, respectively.

<Info>
  The product label is “Grok Imagine 2,” but the request model IDs do not include `2`. For complete requests, parameters, editing examples, and acceptance checks, use the [Grok Imagine 2 API guide](/en/api-capabilities/grok-imagine-image).
</Info>

## Key facts

| Item                       | Current conclusion                                        |
| -------------------------- | --------------------------------------------------------- |
| Standard model             | `grok-imagine-image`, \$0.02/image                        |
| Quality model              | `grok-imagine-image-quality`, \$0.045/image               |
| Aspect ratios              | Currently documented: `1:1`, `16:9`, `9:16`, `4:3`, `3:4` |
| Resolution                 | `1k` and `2k`; currently the same price within each model |
| Batch generation           | `n` up to 10, billed by generated image count             |
| Standard reference editing | `/v1/images/edits`; supports 1–4 references               |
| Quality reference editing  | `/v1/images/edits`; supports 1–2 references               |
| Text-to-image              | `/v1/images/generations`                                  |
| Response format            | `url` or `b64_json`                                       |

<Warning>
  xAI documents additional aspect ratios and publishes its own list prices. LaoZhang API currently commits only to the five ratios in this announcement, and LaoZhang retail pricing is a separate billing contract from xAI list pricing. Reconcile costs against the [LaoZhang console](https://api2.laozhang.ai/account/pricing) and call logs.
</Warning>

<Info>
  “Up to 10 images” is the generated output count controlled by `n`, not a reference-input limit. LaoZhang reference ranges are model-specific: Standard supports 1–4 and Quality supports 1–2.
</Info>

## Who should care

* Developers who need predictable per-image budgets and 1K/2K batch generation;
* Content, advertising, and commerce teams that want multiple candidates in one request;
* Workflows that use multiple references for subject preservation, restyling, or composition and can honor model-specific input limits;
* Teams migrating from GPT-Image, Flux, or another Images API and wanting OpenAI SDK compatibility.

Text, voice, and video calls are not affected. Workflows that require 4K, mask inpainting, arbitrary pixel dimensions, or reproducible `seed` behavior should not assume direct compatibility.

## What to do

<Steps>
  <Step title="Confirm token group and price">
    Open [console model pricing](https://api2.laozhang.ai/account/pricing) and verify that the target token can see the model ID and current rate.
  </Step>

  <Step title="Choose standard or quality">
    Use `grok-imagine-image` for routine assets and three- or four-reference composition. Compare `grok-imagine-image-quality` for final assets or one- to two-reference precision edits.
  </Step>

  <Step title="Complete a one-image acceptance test">
    Test the target ratio and resolution with `n=1`. Inspect actual dimensions, image content, response fields, and the charge before increasing batch size or concurrency.
  </Step>
</Steps>

Minimal text-to-image request:

```bash theme={null}
curl https://api2.laozhang.ai/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-imagine-image",
    "prompt": "A minimalist coffee product poster with warm studio lighting",
    "n": 1,
    "aspect_ratio": "1:1",
    "resolution": "1k"
  }'
```

The canonical guide for text-to-image, Python, reference editing, multi-image composition, migration, and errors is [Grok Imagine 2 API Image Generation and Editing](/en/api-capabilities/grok-imagine-image).

## Frequently asked questions

### What are the correct Grok Imagine 2 model IDs?

Use `grok-imagine-image` and `grok-imagine-image-quality`. Neither contains `2`; do not construct a model ID from the product label.

### Is pricing per request or per image?

It is based on the number of successfully generated images. Current prices are \$0.02/image for `grok-imagine-image` and \$0.045/image for `grok-imagine-image-quality`; `n=10` can therefore create ten image charges. Use the console call log as the billing record.

### Do 1K and 2K cost the same?

Yes under the current LaoZhang launch pricing for a given model. This may change with route or pricing updates, so recheck the console before a large batch.

### Why does the announcement list only five aspect ratios?

They are the current documented LaoZhang support scope. xAI documents more upstream ratios, but upstream capability alone does not prove that every value has been verified and committed on the current LaoZhang gateway.

### Which endpoint accepts reference images?

Use `/v1/images/edits` with `multipart/form-data`: `image` for one file and repeated `image[]` for multi-image edits. `grok-imagine-image` supports 1–4 references, and `grok-imagine-image-quality` supports 1–2. Text-to-image continues to use JSON at `/v1/images/generations`.

### Must existing image workloads migrate now?

No. This is a model addition, not a retirement. Migrate only when predictable per-image cost, multi-image output, or reference editing fits the task, and only after matched-prompt acceptance testing.

## Evidence and related documentation

* [xAI multi-image editing](https://docs.x.ai/developers/model-capabilities/images/multi-image-editing) — upstream multi-image workflow and ordering
* [xAI Grok Imagine Image model](https://docs.x.ai/developers/models/grok-imagine-image) — standard-model image input
* [xAI image generation documentation](https://docs.x.ai/developers/model-capabilities/images/generation) — upstream parameters and response formats
* [xAI models and pricing](https://docs.x.ai/developers/pricing) — xAI list pricing, not LaoZhang API retail pricing
* [Complete Grok Imagine 2 API guide](/en/api-capabilities/grok-imagine-image) — LaoZhang endpoints, code, and acceptance checks
* [Image generation API selection guide](/en/api-capabilities/image-generation-guide) — compare image models
* [LaoZhang API console](https://api2.laozhang.ai/account/pricing) — current groups, prices, and recorded charges