> ## Documentation Index
> Fetch the complete documentation index at: https://aisa.one/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# Image Generation via Chat
> Generate images with wan2.7-image and wan2.7-image-pro. These models route through /v1/chat/completions (not /v1/images/generations) using OpenAI's multimodal chat schema.
Wan 2.7 image models are exposed through the **Chat Completions** endpoint, not the OpenAI-style `/v1/images/generations` path. Send a standard chat request with a multimodal `content` array containing a text prompt, and AIsa returns generated images as `{type: "image"}` parts inside `choices[].message.content[]`.
Looking for Seedream (`seedream-4-5-251128`)? It uses a different route — [`/v1/images/generations`](#). Gemini-compatible `generateContent` requests use [`/v1beta/models/{model}:generateContent`](/api-reference/chat/generatecontent). This page only covers the **Wan 2.7 family**.
## Supported models
| Model | Cost per image | Typical use |
| ------------------ | -------------- | --------------------------------------------------------------- |
| `wan2.7-image` | \$0.030 | Fast, general-purpose image generation |
| `wan2.7-image-pro` | \$0.075 | Higher fidelity; also supports image-to-video via separate flow |
## Request
The request schema is the same `POST /v1/chat/completions` you already use for text — the only differences are which model you pass and how `content` is structured.
**Critical rule:** `messages[].content` must be an **array of typed parts**. Passing a plain string returns `400 invalid_parameter_error` with the message `"Input should be a valid list: messages[*].content"`.
```bash curl theme={null}
curl -sS -X POST "https://api.aisa.one/v1/chat/completions" \
-H "Authorization: Bearer $AISA_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan2.7-image",
"messages": [
{
"role": "user",
"content": [
{ "type": "text", "text": "A cute red panda, ultra-detailed, cinematic lighting" }
]
}
],
"n": 1
}'
```
```python Python theme={null}
from openai import OpenAI
client = OpenAI(base_url="https://api.aisa.one/v1", api_key="sk-aisa-...")
resp = client.chat.completions.create(
model="wan2.7-image",
messages=[
{
"role": "user",
"content": [
{"type": "text", "text": "A cute red panda, ultra-detailed, cinematic lighting"}
],
}
],
n=1,
)
# Pull image URLs out of the response
for choice in resp.choices:
for part in choice.message.content:
if part["type"] == "image":
print(part["image"])
```
```typescript TypeScript theme={null}
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.aisa.one/v1",
apiKey: process.env.AISA_API_KEY,
});
const resp = await client.chat.completions.create({
model: "wan2.7-image",
messages: [
{
role: "user",
content: [
{ type: "text", text: "A cute red panda, ultra-detailed, cinematic lighting" },
] as any,
},
],
n: 1,
});
const urls = resp.choices
.flatMap((c) => (c.message.content as any[]))
.filter((p) => p.type === "image")
.map((p) => p.image);
```
### Request fields
| Field | Type | Required | Notes |
| ------------------------------------ | ------- | --------------------- | ---------------------------------------------------------------------------- |
| `model` | string | yes | `wan2.7-image` or `wan2.7-image-pro` |
| `messages[].role` | string | yes | `user` for the prompt turn |
| `messages[].content` | array | **yes** | Must be an array, not a string |
| `messages[].content[].type` | string | yes | `text` for prompt parts; `image_url` for image-to-image inputs |
| `messages[].content[].text` | string | when `type=text` | The prompt |
| `messages[].content[].image_url.url` | string | when `type=image_url` | Reference image URL |
| `n` | integer | no | Number of images. **Default is 4** for `wan2.7-image`; pass `1` to save cost |
## Response shape
```json theme={null}
{
"id": "chatcmpl-fcc86dfd-...",
"object": "chat.completion",
"created": 1776495713,
"model": "wan2.7-image",
"choices": [
{
"index": 0,
"finish_reason": "stop",
"message": {
"role": "assistant",
"content": [
{ "type": "image", "image": "https://cdn.aisa.one/images/wan2.7/..." }
]
}
}
],
"usage": {
"prompt_tokens": 104,
"completion_tokens": 8,
"total_tokens": 112
}
}
```
* **One choice per image.** If `n=4`, you get 4 entries in `choices`.
* **Every `choice.message.content`** is an array with a single `{ "type": "image", "image": "..." }` part.
* `image` is a short-lived URL (download it soon) or base64 data, depending on your workspace configuration.
* `usage.total_tokens` reflects the small token cost of the request framing — **billing is per-image** at the rate in the table above, not per token.
## Image-to-image
Prepend an `image_url` part to the `content` array and follow it with a text instruction:
```json theme={null}
{
"model": "wan2.7-image-pro",
"messages": [
{
"role": "user",
"content": [
{ "type": "image_url", "image_url": { "url": "https://example.com/reference.jpg" } },
{ "type": "text", "text": "Transform into an oil painting in the style of Van Gogh" }
]
}
],
"n": 1
}
```
## Why the playground shows the Chat Completions path
The playground sends exactly the same `POST /v1/chat/completions` request the standard [OpenAI Chat](/api-reference/chat/createchatcompletion) endpoint uses — only the `model` and `content` shape are tuned for images. Your existing OpenAI-compatible SDK code works without modification; just swap the model and content shape.
## Common 4xx causes
* `400 invalid_parameter_error — Input should be a valid list: messages[*].content` — `content` was passed as a string; wrap in an array of typed parts.
* `400` referencing `messages` — you sent the Gemini-style `contents`/`parts`. Use `messages` with OpenAI multimodal parts for Wan models.
* `404 openai_error` on `/v1/images/generations` — wrong endpoint. Wan models do **not** route through that path.
* `500 model_not_found` — your workspace isn't provisioned for the Wan family. Contact support.
See [Error Codes](/api-reference/errors) and [Rate Limits](/api-reference/rate-limits) for more.
## Related
The same endpoint used for text models.
Gemini-compatible generateContent endpoint.
Agent skill that wraps image + video generation.
## OpenAPI
````yaml openapi/chat-image-generation.json POST /chat/completions
openapi: 3.0.3
info:
title: Image Generation via Chat Completions
version: 1.0.0
description: >-
Wan 2.7 image models (`wan2.7-image`, `wan2.7-image-pro`) are exposed
through the standard `POST /v1/chat/completions` endpoint rather than
`/v1/images/generations`. Send a chat request with a multimodal `content`
array; AIsa returns `choices[].message.content[]` containing one or more
`{type: "image"}` parts.
servers:
- url: https://api.aisa.one/v1
security:
- BearerAuth: []
paths:
/chat/completions:
post:
summary: Generate images via Chat Completions (wan2.7 family)
description: >-
Generate images with the Wan 2.7 image models. Request shape matches
OpenAI Chat Completions with multimodal content; the response contains
`choices[].message.content[]` where each item is `{type: "image", image:
""}`.
**Important:** `messages[].content` must be an **array of typed parts**,
not a plain string. Passing a string returns `400
invalid_parameter_error` with `Input should be a valid list:
messages[*].content`.
By default `wan2.7-image` returns **4 candidate images** (billed per
image). Set `n` to request a different count.
operationId: generateImageViaChat
requestBody:
required: true
content:
application/json:
schema:
type: object
required:
- model
- messages
properties:
model:
type: string
enum:
- wan2.7-image
- wan2.7-image-pro
description: >-
Image-generation model. `wan2.7-image` ($0.030/image) for
standard quality, `wan2.7-image-pro` ($0.075/image) for
higher fidelity.
messages:
type: array
description: >-
Conversation messages. Image prompts go in the last user
message's `content` array as `{type: "text"}` parts.
items:
type: object
required:
- role
- content
properties:
role:
type: string
enum:
- system
- user
- assistant
content:
type: array
description: >-
Multimodal parts. For image generation, include at
least one `{type: "text"}` part with the prompt.
items:
type: object
required:
- type
properties:
type:
type: string
enum:
- text
- image_url
text:
type: string
description: 'Used when `type: "text"`.'
image_url:
type: object
description: >-
Used when `type: "image_url"` (image-to-image
use cases).
properties:
url:
type: string
format: uri
'n':
type: integer
minimum: 1
maximum: 4
default: 4
description: >-
Number of images to generate. `wan2.7-image` returns 4 by
default; pass `1` to save cost.
examples:
basic:
summary: Single-image prompt (n=1)
value:
model: wan2.7-image
messages:
- role: user
content:
- type: text
text: A cute red panda, ultra-detailed, cinematic lighting
'n': 1
four_variants:
summary: Default 4 candidates for selection
value:
model: wan2.7-image
messages:
- role: user
content:
- type: text
text: >-
A futuristic cyberpunk city, neon lights, rainy
night, 8k
pro:
summary: Higher-fidelity pro model
value:
model: wan2.7-image-pro
messages:
- role: user
content:
- type: text
text: >-
Oil painting of a rolling hillside at sunset,
artstation
'n': 1
image_to_image:
summary: Image-to-image (reference + prompt)
value:
model: wan2.7-image-pro
messages:
- role: user
content:
- type: image_url
image_url:
url: https://example.com/reference.jpg
- type: text
text: >-
Transform into an oil painting in the style of Van
Gogh
'n': 1
responses:
'200':
description: >-
Images generated. Returned as Chat Completion with
`message.content[]` image parts.
content:
application/json:
schema:
type: object
properties:
id:
type: string
example: chatcmpl-fcc86dfd-9424-9523-b0bd-cdf07383bee2
object:
type: string
example: chat.completion
created:
type: integer
example: 1776495713
model:
type: string
example: wan2.7-image
choices:
type: array
description: One entry per generated image.
items:
type: object
properties:
index:
type: integer
nullable: true
finish_reason:
type: string
example: stop
message:
type: object
properties:
role:
type: string
example: assistant
content:
type: array
items:
type: object
required:
- type
properties:
type:
type: string
enum:
- image
image:
type: string
description: >-
Short-lived URL to the generated image (or
base64-encoded data, depending on your
account configuration).
usage:
type: object
properties:
prompt_tokens:
type: integer
completion_tokens:
type: integer
total_tokens:
type: integer
example:
id: chatcmpl-fcc86dfd-9424-9523-b0bd-cdf07383bee2
object: chat.completion
created: 1776495713
model: wan2.7-image
choices:
- index: 0
finish_reason: stop
message:
role: assistant
content:
- type: image
image: https://cdn.aisa.one/images/wan2.7/20260418-abc.png
usage:
prompt_tokens: 104
completion_tokens: 8
total_tokens: 112
'400':
description: >-
Invalid request. Most common: `content` passed as a string instead
of an array → `"Input should be a valid list: messages[*].content"`.
'401':
description: Missing or invalid AIsa API key.
'429':
description: Rate limit hit.
'500':
description: Internal error.
'502':
description: Upstream provider unreachable.
components:
securitySchemes:
BearerAuth:
type: http
scheme: bearer
bearerFormat: AISA API Key
````