Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Alibaba Cloud / Qwen MT Image

Qwen MT Image

Translate image text with direction-specific language support and a Chinese/English language-pair requirement.

Image translation
availableClosed weightsImage translation
Official documentation (opens in a new tab)Reviewed 14 Oct 2026
34 of 42 specification fields collected
Source languages11 listedFull language list in specifications
Target languages14 listedFull language list in specifications
Input billingSee ratesAlibaba Model Studio · China (Beijing)
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected
Maximum input tokens
Not collected
Maximum output
Not collected
Input modalities
ImageSource for Input modalities (opens in a new tab)
Output modalities
ImageSource for Output modalities (opens in a new tab)
Reasoning
Not collected
Tool calling
Not supportedSource for Tool calling (opens in a new tab)
Structured output
Not supportedSource for Structured output (opens in a new tab)
Streaming
Not collected
Batch processing
Not supportedSource for Batch processing (opens in a new tab)
Fine-tuning
Not supportedSource for Fine-tuning (opens in a new tab)
Knowledge cutoff
Not collected
Release date
2025-08-22Source for Release date (opens in a new tab)
Parameters
Not collected
Architecture
Not collected
Languages
View all 15 entries
  • Chinese
  • English
  • Korean
  • Japanese
  • Russian
  • Spanish
  • French
  • Portuguese
  • Italian
  • German
  • Vietnamese
  • Malay
  • Thai
  • Indonesian
  • Arabic
Source for Languages (opens in a new tab)

Union of source and target language names. Check direction-specific lists and pairing rules.

Context caching
Not supportedSource for Context caching (opens in a new tab)
Search grounding
Not supportedSource for Search grounding (opens in a new tab)
Prefix completion
Not supportedSource for Prefix completion (opens in a new tab)
Console model playground
Not supportedSource for Console model playground (opens in a new tab)
API aliases
qwen-mt-imageSource for API aliases (opens in a new tab)
Output formats
JPGSource for Output formats (opens in a new tab)
Source languages
Chinese (zh), English (en), Korean (ko), Japanese (ja), Russian (ru), Spanish (es), French (fr), Portuguese (pt), Italian (it), German (de), Vietnamese (vi)Source for Source languages (opens in a new tab)
Target languages
View all 14 entries
  • Chinese (zh)
  • English (en)
  • Korean (ko)
  • Japanese (ja)
  • Russian (ru)
  • Spanish (es)
  • French (fr)
  • Portuguese (pt)
  • Italian (it)
  • Vietnamese (vi)
  • Malay (ms)
  • Thai (th)
  • Indonesian (id)
  • Arabic (ar)
Source for Target languages (opens in a new tab)
Language pair restrictions
At least one side must be Chinese or English. Source/target must differ. German is source-only; Malay, Thai, Indonesian and Arabic are target-only.Source for Language pair restrictions (opens in a new tab)
Automatic language detection
source_lang accepts auto; otherwise language names/codes are case-insensitive. target_lang requires an explicit language.Source for Automatic language detection (opens in a new tab)
Required request fields
image_url, source_lang and target_langSource for Required request fields (opens in a new tab)
Image input limits
One image; up to 100 MB, 15–8,192 pixels per side, aspect ratio 1:10 to 10:1. Public HTTP(S) URL without Chinese characters.Source for Image input limits (opens in a new tab)
Input image formats
JPG, JPEG, PNG, BMP, PNM, PPM, TIFF, WEBPSource for Input image formats (opens in a new tab)
Images per request
1Source for Images per request (opens in a new tab)
Output resolutions
Output preserves source image dimensionsSource for Output resolutions (opens in a new tab)
Output link expiry
JPG result links and asynchronous task IDs expire after 24 hoursSource for Output link expiry (opens in a new tab)

Not a complete input retention or training-use policy.

Translation domain hints
Optional ext.domainHint in English only; up to 200 words recommended for quality. This is not the model input-token limit.Source for Translation domain hints (opens in a new tab)
Sensitive text filtering
Optional case-sensitive exact-match ext.sensitives; up to 50 entries recommended for quality. Words may use a supported source or target language.Source for Sensitive text filtering (opens in a new tab)
Terminology controls
ext.terminologies contains src/tgt term pairs in the selected source/target languages.Source for Terminology controls (opens in a new tab)
Text on image subjects
ext.config.imageSegment=false by default translates all text. true excludes text on detected subjects such as people, products and logos. Legacy skipImgSegment remains accepted; do not assume the two booleans have identical polarity.Source for Text on image subjects (opens in a new tab)
Asynchronous generation
SupportedSource for Asynchronous generation (opens in a new tab)
Synchronous generation
Not supportedSource for Synchronous generation (opens in a new tab)

The shared request schema names both models, but the explicit header rule allows synchronous calls only for 2.0.

Request mode
Asynchronous only: X-DashScope-Async: enable is required; poll task status.Source for Request mode (opens in a new tab)
Output billing
Per successfully generated translated image. Input is free; failures do not incur fees.Source for Output billing (opens in a new tab)
Usage metering
Successful translated images × native image rateSource for Usage metering (opens in a new tab)
Published availability regions
China (Beijing)Source for Published availability regions (opens in a new tab)

API overview confirms the same regional availability as pricing.

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Alibaba Model Studio · China (Beijing)

Provider source (opens in a new tab)
API model ID
qwen-mt-image
Access
POST https://{WorkspaceId}.cn-beijing.maas.aliyuncs.com/api/v1/services/aigc/image2image/image-synthesis. Asynchronous only: X-DashScope-Async: enable is required; poll task status. Async results: GET /api/v1/tasks/{task_id}. Use the matching regional workspace and API key.
Regions
China (Beijing)
Tool calling
Not supportedSource for Tool calling (opens in a new tab)
Structured output
Not supportedSource for Structured output (opens in a new tab)
Search grounding
Not supportedSource for Search grounding (opens in a new tab)
Prefix completion
Not supportedSource for Prefix completion (opens in a new tab)
Context caching
Not supportedSource for Context caching (opens in a new tab)
Batch processing
Not supportedSource for Batch processing (opens in a new tab)
Fine-tuning
Not supportedSource for Fine-tuning (opens in a new tab)
Request mode
Asynchronous only: X-DashScope-Async: enable is required; poll task status.Source for Request mode (opens in a new tab)
Synchronous generation
Not supportedSource for Synchronous generation (opens in a new tab)

The shared request schema names both models, but the explicit header rule allows synchronous calls only for 2.0.

Task submission rate
1 request per secondSource for Task submission rate (opens in a new tab)

1 Times/s

Concurrent requests
Up to 2 asynchronous tasksSource for Concurrent requests (opens in a new tab)

The quota page also labels synchronous concurrency unlimited for the old model; its API explicitly requires asynchronous calls, so this is not treated as sync support.

Published request quota
60 requests per minuteSource for Published request quota (opens in a new tab)
Quota scope
Primary account aggregate across RAM users, workspaces and API keys; per-model quotaSource for Quota scope (opens in a new tab)
Free usage allowance
No free quotaSource for Free usage allowance (opens in a new tab)
Regional route evidence
Image translation path from its API; regional workspace host from the provider region guide.Source for Regional route evidence (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 14 Oct 2026
CategoryUsageRateUnit
Image translationSuccessful translated image1$0.000431image

Rate conditions

  1. Output-only native USD billing. Input is free and failed requests do not incur fees. JPG output preserves source dimensions. No text-token or per-question estimate.
Pricing source (opens in a new tab)

Data & enterprise terms

Result links expire after 24 hours. Full input retention, training use and enterprise policy remain unverified.

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Classification inference: this separately documented native image-translation service has no exact public checkpoint in the reviewed official Qwen MT search. API branding and an empty inventory query do not prove private lineage. Text-translation and image-generation weights are not assumed equivalent.

Classification evidence (opens in a new tab)

Lifecycle dates

Platform release · China (Beijing)22 Aug 2025Source (opens in a new tab)

Sources reviewed