Last verified
RETIRING OCT 10 2026FRONTIER QWEN252K CONTEXTBATCH -50%TEXT + VISION

Qwen3 Max API Pricing

Retiring October 10, 2026: Alibaba has announced that qwen3-max goes offline on that date, after which calls fail. Alibaba recommends Qwen3.7 Max instead. Qwen3 Max is Qwen's proprietary flagship. The International deployment row lists $1.20/M input and $6.00/M output for 0-32K input tokens. Alibaba also publishes lower Global/Chinese Mainland rows, so region matters.

Input - per 1M tokens
$1.20/M
Source alibabacloud.com flat
Output - per 1M tokens
$6.00/M
Context 252K flat
Cache N/A
$1.20/M
Cache vendor row not listed
Effective - agentic blend
$1.58/M
92/8 split - no cache
§ 01 / TERMINAL

Run the numbers.

Live calculator pre-loaded with Qwen3 Max rates. Tweak spend, output mix, or cache assumptions to compare it with sibling models.

$ /mo
Workload split
Tokens you can process
—
Words equivalent (English)
—
Effective rate
—
Open full calculator (all models · share URL · CSV) →
§ 02 / SCENARIOS

Real-world presets.

VISION

Invoice extraction

$0.012/doc
6,000 in - 800 out~8,333 units/$100
CHATBOT

Support assistant

$0.0066/turn
2,500 in - 600 out~15,151 units/$100
RAG

Knowledge base answer

$0.017/query
9,000 in - 1,000 out~5,952 units/$100
BULK

Image QA review

$0.006/item
3,000 in - 400 out~16,666 units/$100
§ 03 / TOKENIZER

Paste text. See tokens. See cost.

What's a token →
Calibrated · measured on the vendor's tokenizer · 2026-06-10 Auto-counts as you type

Counts use a chars-per-token calibration measured on the vendor's own published tokenizer (Qwen/Qwen3.5-397B-A17B, 2026-06-10). English prose is typically within a few percent; code and non-Latin scripts tokenize heavier. For billing-exact counts use the vendor's count-tokens API.

Characters 369
Words 58
Tokens (estimated) 71 tokens
Cost as input · uncached $0.00009 USD
Cost as output · uncached $0.00043 USD
§ 04 / SHELF

Up against the shelf.

All models →
Model Input /M Output /M Effective blended Context Best for
Qwen3 Max Current $1.20 $6.00 $1.58 agentic 92/8 252K Frontier Qwen reasoning
Qwen 3.5 Plus $0.40 $2.40 $0.56 −2.8× 256K General Qwen production workloads
Qwen 3.5 Flash $0.10 $0.40 $0.124 −13× 1M Bulk chat and long-context RAG
Qwen3 VL Plus $0.20 $1.60 $0.312 −5.1× 256K Vision and document understanding
Qwen3 VL Flash $0.05 $0.40 $0.078 deprecated · −20× 256K Vision and document understanding
GPT-5.4 mini $0.75 cache $0.075 $4.50 $0.541 −2.9× 400K Subagents and lightweight coding
Gemini 3.8 Flash $0.75 cache $0.075 $3.75 $0.481 −3.3× · promo 1M Mainstream multimodal work

Frequently asked.

Practical pricing questions, separated from calculator assumptions.

Q · 01 When does Qwen3 Max stop working? +
Alibaba announced on June 8, 2026 that qwen3-max goes offline on October 10, 2026 (00:00, UTC+8) in the Beijing and Singapore regions; after that, calls fail. Its notice recommends Qwen3.7 Max. Until then the model bills at the rates on this page.
Q · 02 What is Qwen3 Max priced at? +
Qwen3 Max is shown here at the International deployment row: $1.20/M input and $6.00/M output for 0-32K input tokens. Alibaba's same pricing page also publishes lower Global and Chinese Mainland deployment rows.
Q · 03 Does this page include higher context pricing tiers? +
Alibaba publishes tiered pricing by input length and by deployment mode. This page uses the International 0-32K row; longer prompts and other regions can have different prices.
Q · 04 Is prompt caching priced separately? +
No separate cache-read price is published for this row, so the calculator treats cached input as $1.2/M.
Q · 05 How is the effective price calculated? +
AI//COST uses the same 92/8 agentic blend everywhere. For Qwen3 Max, that gives $1.58/M with only documented cache discounts included.
Q · 06 Is there a batch discount? +
Alibaba lists Batch Invocation at 50% off for supported Qwen rows.
Q · 07 Are regional prices different? +
Yes. Alibaba Cloud publishes separate International, Global, US, EU, China (Hong Kong), and Chinese Mainland deployment sections. AI//COST uses the International / Singapore baseline for this queue unless a page explicitly says otherwise.
Y
Reviewed by Yaroslav Vikhariev Founder · AI//COST · Prices checked by hand against alibabacloud.com · Last verified Jun 8, 2026
Methodology Report a correction More by Y.V.