Alibaba Cloud: Qwen-Doc-Turbo

Qwen Doc Turbo focuses on working with document text. Its long context is useful for locating details across a lengthy file, comparing passages, and answering questions that depend on more than a short excerpt.

Compare models

qwen-doc-turbo

AvailableAlibaba CloudChat

Input / Output: $0.087 / $0.144
Context: 254K
Max output: 8K
Modalities: Chat
Capabilities: Prompt Cache

About Qwen-Doc-Turbo

Qwen-Doc-Turbo is Alibaba's Qwen model for document mining: pulling structured information out of long files. Alibaba positions it for information extraction, content moderation, data classification and summarization, and says it returns JSON more accurately than its general chat models. Unlike Qwen Plus or Qwen Flash, it is a task-specific extractor rather than an all-round assistant, and it keeps output short.

Where it works well

Extracting fields from long reports, contracts and spreadsheets into JSON, which is the job Alibaba built it for.
Classifying and moderating passages in bulk, returning a label per item instead of free-form prose.
Summarizing a lengthy file in one pass without splitting it into chunks first.
Prompt caching reduces repeat work when many questions target the same document.

When to choose another model

Text input only with no tool calling, so scanned pages need OCR first.
Answers are capped at a short length, so rewriting or drafting a long document needs a model with a larger output budget.
For open-ended chat, coding or multi-step reasoning, Qwen Plus or Qwen3.8 Max is a better fit.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Text to textMessages API

POST/v1/messages

API format:

curl https://api.tokenlab.sh/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: sk-xxx" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "qwen-doc-turbo",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Default

per 1M tokens

Official price: Input $0.087 / Output $0.144 / Cache read $0.0174 / Cache write $0.1087 / Text Input $0.087 / Text Output $0.144
Verified price: —
Discount: —

Prompt cache pricing

Cache read

Official price: $0.0174
Verified price: —
Discount: —

Cache write

Official price: $0.1087
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Default	Input $0.087 / Output $0.144 / Cache read $0.0174 / Cache write $0.1087 / Text Input $0.087 / Text Output $0.144	—	—
Prompt cache pricing
Cache read	$0.0174	—	—
Cache write	$0.1087	—	—

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Nov 1, 2026, 10:24 AM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Qwen-Doc-Turbo in Console with a prompt ready to edit or send.

Help me try qwen-doc-turbo with a short message at /v1/messages. Show the reply, latency, and cost.

Use cases

Invoice and contract field extraction
Feed supplier contracts or invoices and ask for party names, dates, amounts and renewal terms as a fixed JSON object, so results load straight into a database.
Bulk document tagging
Label a backlog of reports or tickets by topic, urgency or department, returning one category per document for filtering later.
Compliance screening
Scan long submissions for passages that break a stated policy and return the offending quotes with a reason for each.
Digest generation
Turn a long meeting transcript or filing into a fixed-format summary with decisions, owners and open questions.

Prompt examples

Extract vendor, effective date, termination notice period and payment terms from this contract. Return JSON only, with null for anything not stated.

Classify each of the following 30 support emails as billing, bug, feature request or other. Return a JSON array of ids and labels.

Summarize this annual report in five bullet points and list every risk factor the company names.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Qwen-Doc-Turbo used for?

It is built for document mining: extracting structured data, classifying, moderating and summarizing long text. Alibaba highlights consistent JSON output as its main advantage over general chat models. Use it when you have many documents and a fixed schema, not for open conversation.

Is Qwen-Doc-Turbo better than Qwen Plus for extraction?

For fixed-schema extraction from long files, Alibaba positions Doc-Turbo as the more accurate and consistent choice. Qwen Plus is the better pick when the same request also needs analysis, writing or free-form dialogue. Test both on a sample of your own documents before committing.

Can Qwen-Doc-Turbo read PDFs or images?

It takes text, so convert PDFs and scans to text before sending them. Run OCR first for scanned pages, then pass the extracted text in the prompt.

Does Qwen-Doc-Turbo support tool calling or JSON schema mode?

No. Tool calling and response-schema mode are not offered for this model. Ask for JSON in the prompt, show an example object, and validate the result in your code.

Does Qwen-Doc-Turbo support prompt caching?

Yes. Caching helps when you repeat a long document prefix and vary only the question, such as asking ten different extraction questions about one filing.

How much does Qwen-Doc-Turbo cost?

On TokenLab, Qwen-Doc-Turbo costs Input $0.087 / Output $0.144 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

What are the context window and output limit of Qwen-Doc-Turbo?

Qwen-Doc-Turbo accepts up to 253,952 tokens of context and returns up to 8,192 tokens in one response.

Which endpoint should Qwen-Doc-Turbo use?

Use https://api.tokenlab.sh/v1/messages for Qwen-Doc-Turbo. The request example below shows the matching code shape.

Which operations does Qwen-Doc-Turbo support?

Qwen-Doc-Turbo supports Text to text. Select an operation above to see its endpoint and request example.