> ## Documentation Index
> Fetch the complete documentation index at: https://api-tools.memories.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Qwen Image
> Generate chat completions using Qwen ILM model with image inputs.
**Product**: Visual Intelligence — Image Model APIs (atomic)
**Use case**: Call an Image Language Model (Gemini / GPT / Nova / Qwen) directly with your own prompt. Full control over model and parameters.
**Host**: `https://mavi-backend.memories.ai/serve/api/v2`
**Auth**: `Authorization: sk-mavi-...` (no `Bearer` prefix)
This endpoint allows you to generate chat completions with image inputs using Qwen ILM model.
**Endpoint:** `POST https://mavi-backend.memories.ai/serve/api/v2/iu/chat/completions`
Image Understanding (ILM) endpoints use the `/iu` path prefix. Video Understanding (VLM) endpoints use `/vu` instead.
## Supported Models
All models require the `qwen:` prefix when used in the `model` parameter (e.g., `qwen:qwen3-vl-plus`).
### Qwen 3 VL Plus (Premium - Tiered Pricing)
| Model | Input Price | Output Price |
| ---------------------------- | ---------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| **qwen3-vl-plus** | \$0.00021/1K (≤32K), \$0.00031/1K (32K-128K), \$0.00063/1K (128K-256K) | \$0.00168/1K (≤32K), \$0.00252/1K (32K-128K), \$0.00503/1K (128K-256K) |
| **qwen3-vl-plus-2025-12-19** | \$0.00021/1K (≤32K), \$0.00031/1K (32K-128K), \$0.00063/1K (128K-256K) | \$0.00168/1K (≤32K), \$0.00252/1K (32K-128K), \$0.00503/1K (128K-256K) |
| **qwen3-vl-plus-2025-09-23** | \$0.00021/1K (≤32K), \$0.00031/1K (32K-128K), \$0.00063/1K (128K-256K) | \$0.00168/1K (≤32K), \$0.00252/1K (32K-128K), \$0.00503/1K (128K-256K) |
### Qwen 3 VL Flash (Cost-Effective - Tiered Pricing)
| Model | Input Price | Output Price |
| ----------------------------- | ------------------------------------------------------------------------- | ---------------------------------------------------------------------- |
| **qwen3-vl-flash** | \$0.000052/1K (≤32K), \$0.000079/1K (32K-128K), \$0.000126/1K (128K-256K) | \$0.00042/1K (≤32K), \$0.00063/1K (32K-128K), \$0.00101/1K (128K-256K) |
| **qwen3-vl-flash-2025-10-15** | \$0.000052/1K (≤32K), \$0.000079/1K (32K-128K), \$0.000126/1K (128K-256K) | \$0.00042/1K (≤32K), \$0.00063/1K (32K-128K), \$0.00101/1K (128K-256K) |
### Qwen VL Max (Legacy)
| Model | Input Price | Output Price |
| -------------------------- | ------------------- | ------------------- |
| **qwen-vl-max** | \$0.00084/1K tokens | \$0.00335/1K tokens |
| **qwen-vl-max-latest** | \$0.00084/1K tokens | \$0.00335/1K tokens |
| **qwen-vl-max-2025-08-13** | \$0.00084/1K tokens | \$0.00335/1K tokens |
| **qwen-vl-max-2025-04-08** | \$0.00084/1K tokens | \$0.00335/1K tokens |
### Qwen VL Plus (Legacy)
| Model | Input Price | Output Price |
| --------------------------- | ------------------- | ------------------- |
| **qwen-vl-plus** | \$0.00022/1K tokens | \$0.00066/1K tokens |
| **qwen-vl-plus-latest** | \$0.00022/1K tokens | \$0.00066/1K tokens |
| **qwen-vl-plus-2025-08-15** | \$0.00022/1K tokens | \$0.00066/1K tokens |
| **qwen-vl-plus-2025-05-07** | \$0.00022/1K tokens | \$0.00066/1K tokens |
| **qwen-vl-plus-2025-01-25** | \$0.00022/1K tokens | \$0.00066/1K tokens |
**Pricing Notes:**
* All prices are per 1,000 tokens and converted from CNY at exchange rate of 7.0
* **Tip:** To compare with other providers (who price per 1M tokens), multiply these prices by 1,000. For example, \$0.00021/1K = \$0.21/1M tokens
* Qwen 3 VL models support tiered pricing: ≤32K, 32K-128K, 128K-256K token ranges
* Legacy models (Qwen VL Max/Plus) use flat-rate pricing
* Remember to include the `qwen:` prefix: `"model": "qwen:qwen3-vl-plus"`
### Request Body
| Parameter | Type | Required | Default | Description |
| ------------------ | ----------------------- | -------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| model | string | Yes | - | The model to use (e.g., `qwen:qwen3-vl-plus`) |
| messages | array | Yes | - | Array of message objects. Each message contains:
- `role`: Role type, values: `system`, `user`
- `content`: Message content, can be a string or array. Array items can contain:
- `image`: Image URL or base64 encoded image
- `text`: Text content |
| temperature | number | No | 0.7 | Controls randomness: 0.0-2.0, higher = more random |
| max\_tokens | integer | No | 1024 | Maximum number of tokens to generate |
| top\_p | number | No | 0.9 | Nucleus sampling: 0.0-1.0, consider tokens with top\_p probability mass |
| frequency\_penalty | number | No | 0.0 | Reduces repetition of frequent tokens: -2.0 to 2.0 |
| presence\_penalty | number | No | 0.0 | Increases likelihood of new topics: -2.0 to 2.0 |
| n | integer | No | 1 | Number of completions to generate |
| stream | boolean | No | false | Whether to stream the response |
| stop | string \| array \| null | No | null | Stop sequences. Can be a string, array of strings, or null |
| extra\_body | object | No | - | Additional body parameters. Contains:
- `metadata`: Metadata object
- `enable_thinking`: Boolean to enable thinking mode
- `thinking_budget`: Integer value for thinking budget
- `response_format`: Response format configuration
- `type`: Format type (`json_schema`)
- `json_schema`: JSON schema object
- `name`: Schema name
- `schema`: JSON schema definition |
### Code Example
```python Python theme={null}
from openai import OpenAI
client = OpenAI(
api_key="sk-mavi-...",
base_url="https://mavi-backend.memories.ai/serve/api/v2/iu"
)
resp = client.chat.completions.create(
model="qwen:qwen3-vl-plus",
messages=[
{"role": "system", "content": "You are a multimodal assistant. Keep your answers concise."},
{
"role": "user",
"content": [
{
"image": "https://storage.googleapis.com/memories-test-data/gun5.png" # url or base64
},
{"text": "What is the content of this image?"}
]
}
],
temperature=0.7, # Controls randomness: 0.0-2.0, higher = more random
max_tokens=1024, # Maximum number of tokens to generate
top_p=0.9, # Nucleus sampling: 0.0-1.0, consider tokens with top_p probability mass
frequency_penalty=0.0, # -2.0 to 2.0, reduces repetition of frequent tokens
presence_penalty=0.0, # -2.0 to 2.0, increases likelihood of new topics
n=1, # Number of completions to generate
stream=False, # Whether to stream the response
stop=None, # Stop sequences (list of strings)
extra_body={
"metadata": {
"enable_thinking": True,
"thinking_budget": 1024,
"response_format": {
"type": "json_schema",
"json_schema": {
"name": "answer",
"schema": {
"type": "object",
"properties": {
"result": {"type": "string"}
},
"required": ["result"]
}
}
}
}
}
)
print(resp)
```
### Response
Returns the chat completion response with structured output.
```json theme={null}
{
"id": "6612267a-d08f-9ea0-a254-7bcea6339f49",
"object": "completion",
"model": "qwen3-vl-plus",
"created_at": 1767094289,
"status": "completed",
"choices": [
{
"text": "This image is from the game Half-Life 2, showing a game scene from a first-person perspective.\n\nIn the scene, the player holds a \"Gravity Gun\" in each hand, aiming at the wall ahead. A yellow \"Mandatory Reminder\" sign hangs on the wall, printed with the U.S. national emblem pattern and inscribed with \"AMENDMENT 35\" (35th Amendment), with text below reading: \"No person shall be found with fewer than thirteen times their own body weight of federally provisioned munitions.\" - This is clearly a fictional absurd law within the game.\n\nThe bottom left corner of the screen shows the player's \"HEALTH\" and \"SUIT\" both at 100. The bottom right corner displays ammunition information \"IDC ANYMORE AMMO\", suggesting ample or infinite ammunition.\n\nThe entire scene is full of dark humor and dystopian style, one of the iconic scenes from Half-Life 2, often used by players as memes or screenshots for sharing.",
"index": 0
}
],
"usage": {
"input_tokens": 235,
"output_tokens": 229,
"total_tokens": 464
},
"meta": {
"provider": "qwen",
"provider_model": "qwen3-vl-plus"
}
}
```
### Response Parameters
| Parameter | Type | Description |
| -------------------- | ------- | ------------------------------------------------- |
| id | string | Unique identifier for the completion |
| object | string | Object type, always "completion" |
| model | string | The model used for the completion |
| created\_at | integer | Unix timestamp of when the completion was created |
| status | string | Status of the completion (e.g., "completed") |
| choices | array | Array of completion choices |
| choices\[].text | string | Text content of the completion |
| choices\[].index | integer | Index of the choice in the choices array |
| usage | object | Token usage information |
| usage.input\_tokens | integer | Number of input tokens used |
| usage.output\_tokens | integer | Number of output tokens generated |
| usage.total\_tokens | integer | Total number of tokens used |
| meta | object | Metadata about the completion |
| meta.provider | string | Provider name (e.g., "qwen") |
| meta.provider\_model | string | Provider-specific model name |
## OpenAPI
````yaml visual-intelligence/qwen/openapi.json POST /iu/chat/completions
openapi: 3.1.0
info:
title: Qwen API Reference
description: REST APIs for Qwen ILM and VLM models
version: v1.0.0
servers:
- url: https://mavi-backend.memories.ai/serve/api/v2
security:
- ApiKeyAuth: []
paths:
/iu/chat/completions:
post:
summary: Chat Completions qwen ilm
description: >-
Generate chat completions using Qwen ILM or VLM models with image and
video inputs.
operationId: chat_completions_qwen_ilm
requestBody:
required: true
content:
application/json:
schema:
type: object
properties:
model:
type: string
description: The model to use (e.g., qwen:qwen3-vl-plus)
example: qwen:qwen3-vl-plus
messages:
type: array
description: Array of message objects
items:
type: object
properties:
role:
type: string
enum:
- system
- user
content:
oneOf:
- type: string
- type: array
items:
oneOf:
- type: object
name: image
properties:
image:
type: string
description: URL or base64 encoded image
- type: object
name: image
properties:
text:
type: string
required:
- text
required:
- role
- content
temperature:
type: number
description: 'Controls randomness: 0.0-2.0, higher = more random'
minimum: 0
maximum: 2
default: 0.7
max_tokens:
type: integer
description: Maximum number of tokens to generate
default: 1024
top_p:
type: number
description: 'Nucleus sampling: 0.0-1.0'
minimum: 0
maximum: 1
default: 0.9
frequency_penalty:
type: number
description: 'Reduces repetition of frequent tokens: -2.0 to 2.0'
minimum: -2
maximum: 2
default: 0
presence_penalty:
type: number
description: 'Increases likelihood of new topics: -2.0 to 2.0'
minimum: -2
maximum: 2
default: 0
'n':
type: integer
description: Number of completions to generate
default: 1
stream:
type: boolean
description: Whether to stream the response
default: false
stop:
oneOf:
- type: string
- type: array
items:
type: string
- type: 'null'
description: Stop sequences
extra_body:
type: object
properties:
metadata:
type: object
properties:
enable_thinking:
type: boolean
thinking_budget:
type: integer
response_format:
type: object
properties:
type:
type: string
enum:
- json_schema
json_schema:
type: object
properties:
name:
type: string
schema:
type: object
properties:
type:
type: string
enum:
- object
properties:
type: object
required:
type: array
items:
type: string
required:
- model
- messages
responses:
'200':
description: Chat completion response with structured output
content:
application/json:
schema:
type: object
properties:
id:
type: string
example: 6612267a-d08f-9ea0-a254-7bcea6339f49
description: Unique identifier for the completion
object:
type: string
example: completion
description: Object type, always 'completion'
model:
type: string
example: qwen3-vl-plus
description: The model used for the completion
created_at:
type: integer
example: 1767094289
description: Unix timestamp of when the completion was created
status:
type: string
example: completed
description: Status of the completion
choices:
type: array
items:
type: object
properties:
text:
type: string
description: Text content of the completion
example: >-
This image is from the game Half-Life 2, showing a
game scene from a first-person perspective.
index:
type: integer
description: Index of the choice in the choices array
example: 0
required:
- text
- index
usage:
type: object
properties:
input_tokens:
type: integer
description: Number of input tokens used
example: 235
output_tokens:
type: integer
description: Number of output tokens generated
example: 229
total_tokens:
type: integer
description: Total number of tokens used
example: 464
required:
- input_tokens
- output_tokens
- total_tokens
meta:
type: object
properties:
provider:
type: string
description: Provider name
example: qwen
provider_model:
type: string
description: Provider-specific model name
example: qwen3-vl-plus
required:
- provider
- provider_model
required:
- id
- object
- model
- created_at
- status
- choices
- usage
- meta
components:
securitySchemes:
ApiKeyAuth:
type: apiKey
in: header
name: Authorization
````