All Products
Search
Document Center

Alibaba Cloud Model Studio:Model Invocation Pricing

Last Updated:Nov 18, 2026

Model invocation uses the pay-as-you-go billing method by default, covering the billing rules and prices of various models.

NoteThis page shows only the Original prices of model invocation. Go to the Model Studio console to view promotional offers.

NoteSome models support context caching (Explicit Cache and implicit cache). Input tokens that hit the cache and tokens used to create an Explicit Cache are billed at unit prices different from standard input (for example, Explicit Cache creation is billed at 125% of the standard input unit price, and cache hits are billed at 10%). The input unit prices in the price tables of this page do not include cache prices. For the billing rules, unit price discounts, and supported models of caching, see Context cache.

Tiered billing rules

Some Model Studio models adopt tiered billing. The unit price depends on the total number of input tokens in a single request. All tokens of that request are billed at the unit price of the corresponding tier.

In the billing ranges, K indicates 1,000 and M indicates 1,000,000. For example, 128K means 128,000 tokens, 256K means 256,000 tokens, and 1M means 1,000,000 tokens.

For example, a model has two billing ranges: 0 < Token ≤ 32K and 32K < Token ≤ 128K. If the input is 100K tokens, since the value falls in the second range (32K < 100K ≤ 128K), all tokens are billed at the unit price of the second tier.

Text Generation - Qwen

Qwen Max

Billing rule: Billing is based on input tokens and output tokens.

Factors that affect billing: If the model supports Batch calls, the unit prices of both input and output tokens are billed at 50% of the real-time inference price; if the model supports context caching, only input tokens enjoy a discount. The two cannot take effect at the same time.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen4-max

Cache unit price (per million tokens)

International

Non-thinking and thinking modes

0<Token≤1M

$1.5

$4.5

1 million Tokens

qwen3.8-max

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2

$6

1 million Tokens

qwen3.8-max-0902

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2

$6

None

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2.5

$7.5

1 million Tokens

qwen3.7-max-2026-06-08

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2.5

$7.5

None

qwen3.7-max-2026-05-20

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2.5

$7.5

None

qwen3.7-max-preview

Current capabilities are equivalent to qwen3.7-max-2026-05-17

International

Thinking mode only

0<Token≤1M

$2.5

$7.5

None

qwen3.7-max-2026-05-17

International

Thinking mode only

0<Token≤1M

$2.5

$7.5

None

More Models
Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen-max

Batch calls at half price

International

Non-thinking mode only

No tiered pricing

$1.6

$6.4

None

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.875$3$0.075

China (Beijing)

Model IDPatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen4-max

Cache unit price (per million tokens)

Non-thinking and thinking modes

0<Token≤1M

$1.413

$4.24

qwen3.8-max

Batch calls at half price

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-max-0902

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Batch calls at half price

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-06-08

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-05-20

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

More Models

Model ID

Service Deployment Scope

Pattern

Input tokens per request

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-max

the Chinese mainland

Non-thinking mode only

No tiered pricing

$0.345

$1.377

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.767$2.827$0.071

qwen4-max — Batch unit prices (per million tokens)

APIInputOutput
Batch File$0.707$2.12
Batch Chat (limited-time 50% off)Original $1.413Original $4.24

Hong Kong (China)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen4-max

Cache unit price (per million tokens)

Global

Non-thinking and thinking modes

0<Token≤1M

$1.413

$4.24

qwen3.8-max

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-max-0902

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-06-08

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.767$2.827$0.071

Germany (Frankfurt)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen4-max

Cache unit price (per million tokens)

Global

Non-thinking and thinking modes

0<Token≤1M

$1.413

$4.24

qwen3.8-max

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-max-0902

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-06-08

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.767$2.827$0.071

United States (Virginia)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen4-max

Cache unit price (per million tokens)

Global

Non-thinking and thinking modes

0<Token≤1M

$1.413

$4.24

qwen3.8-max

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-max

Context cache enjoys a discount

United States

Non-thinking and thinking modes

0<Token≤1M

$2

$6

qwen3.8-max-0902

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max

Context cache enjoys a discount

United States

Non-thinking and thinking modes

0<Token≤1M

$2.5

$7.5

qwen3.7-max-2026-06-08

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.767$2.827$0.071

Japan (Tokyo)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen4-max

Cache unit price (per million tokens)

Global

Non-thinking and thinking modes

0<Token≤1M

$1.413

$4.24

qwen3.8-max

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-max-0902

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.7-max

Current capabilities are equivalent to qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen3.7-max-2026-05-20

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤1M

$1.65

$4.951

qwen4-max — Cache unit prices (per million tokens)

Cache creation (5m)Cache creation (1h)Explicit/implicit cache hit
$1.767$2.827$0.071

Qwen Plus

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

International

0<Token≤256K

Original$0.4 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

1 million Tokens

256K<Token≤1M

Original$1.2 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

International

0<Token≤256K

$0.4

$1.6

$1.6

None

256K<Token≤1M

$1.2

$4.8

$4.8

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

International

0<Token≤256K

$0.5

$3

$3

None

256K<Token≤1M

$2

$6

$6

qwen3.6-plus-2026-04-02

International

0<Token≤256K

$0.5

$3

$3

None

256K<Token≤1M

$2

$6

$6

qwen3.5-plus

Current capabilities are equivalent to qwen3.5-plus-2026-02-15

International

0<Token≤256K

$0.4

$2.4

$2.4

None

256K<Token≤1M

$0.5

$3

$3

qwen3.5-plus-2026-04-20

International

0<Token≤256K

$0.4

$2.4

$2.4

None

256K<Token≤1M

$0.5

$3

$3

qwen3.5-plus-2026-02-15

International

0<Token≤256K

$0.4

$2.4

$2.4

None

256K<Token≤1M

$0.5

$3

$3

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

International

0<Token≤256K

$0.4

$1.2

$4

None

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-latest

International

0<Token≤256K

$0.4

$1.2

$4

None

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-2025-12-01

International

0<Token≤256K

$0.4

$1.2

$4

None

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-2025-07-14

International

No tiered pricing

$0.4

$1.2

$4

None

qwen-plus-2025-04-28

International

No tiered pricing

$0.4

$1.2

$4

None

qwen-plus-2025-01-25

International

No tiered pricing

$0.4

$1.2

1 million Tokens

China (Beijing)

Model IDInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

0<Token≤256K

Original$0.276 (20% off for a limited time)

Original$1.101 (20% off for a limited time)

Original$1.101 (20% off for a limited time)

256K<Token≤1M

Original$0.826 (20% off for a limited time)

Original$3.301 (20% off for a limited time)

Original$3.301 (20% off for a limited time)

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

0<Token≤256K

$0.276

$1.101

$1.101

256K<Token≤1M

$0.826

$3.301

$3.301

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.6-plus-2026-04-02

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.5-plus

Current capabilities are equivalent to qwen3.5-plus-2026-02-15

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen3.5-plus-2026-04-20

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen3.5-plus-2026-02-15

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus-latest

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus-2025-12-01

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus-2025-07-14

No tiered pricing

$0.115

$0.287

$1.147

qwen-plus-2025-04-28

No tiered pricing

$0.115

$0.287

$1.147

More Models

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-plus-2025-01-25

the Chinese mainland

No tiered pricing

$0.115

$0.287

qwen-plus-2025-01-12

the Chinese mainland

No tiered pricing

$0.115

$0.287

qwen-plus-2024-12-20

the Chinese mainland

No tiered pricing

$0.115

$0.287

Hong Kong (China)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

Original$0.276 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

256K<Token≤1M

Original$0.826 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

$0.276

$1.101

$1.101

256K<Token≤1M

$0.826

$3.301

$3.301

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

Hong Kong (China)

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-2025-12-01

Hong Kong (China)

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

Germany (Frankfurt)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

Original$0.276 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

256K<Token≤1M

Original$0.826 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

$0.276

$1.101

$1.101

256K<Token≤1M

$0.826

$3.301

$3.301

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.5-plus

Current capabilities are equivalent to qwen3.5-plus-2026-02-15

Global

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen3.5-plus-2026-02-15

Global

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

Global

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

European Union

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-2025-12-01

Global

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus-2025-12-01

European Union

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

United States (Virginia)

NoteThe following table shows the original prices. Some models have limited-time off-peak discounts (see the label next to the price). Off-peak hours are 22:00 to 8:00 the next day in UTC+8, and all other hours are peak hours, based on the billing time.

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

Original$0.276 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

256K<Token≤1M

Original$0.826 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

qwen3.7-plus

Context cache enjoys a discount

United States

0<Token≤256K

Original$0.4 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

256K<Token≤1M

Original$1.2 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

$0.276

$1.101

$1.101

256K<Token≤1M

$0.826

$3.301

$3.301

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.276

$1.651

$1.651

256K<Token≤1M

$1.101

$6.602

$6.602

qwen3.5-plus

Current capabilities are equivalent to qwen3.5-plus-2026-02-15

Global

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen3.5-plus-2026-02-15

Global

0<Token≤128K

$0.115

$0.688

$0.688

128K<Token≤256K

$0.287

$1.72

$1.72

256K<Token≤1M

$0.573

$3.44

$3.44

qwen-plus

Current capabilities are equivalent to qwen-plus-2025-12-01

Global

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus

United States

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

qwen-plus-2025-12-01

Global

0<Token≤128K

$0.115

$0.287

$1.147

128K<Token≤256K

$0.345

$2.868

$3.441

256K<Token≤1M

$0.689

$6.881

$9.175

qwen-plus-2025-12-01

United States

0<Token≤256K

$0.4

$1.2

$4

256K<Token≤1M

$1.2

$3.6

$12

Japan (Tokyo)

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)
Non-thinking modeThinking mode (chain-of-thought + response)

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Japan

0<Token≤256K

Original$0.4 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

Original$1.6 (20% off for a limited time)

256K<Token≤1M

Original$1.2 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

Original$4.8 (20% off for a limited time)

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Japan

0<Token≤256K

$0.4

$1.6

$1.6

256K<Token≤1M

$1.2

$4.8

$4.8

qwen3.7-plus

Current capabilities are equivalent to qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

Original$0.276 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$1.101 (limited-time 60% off for off-peak usage) 20% off during peak hours

256K<Token≤1M

Original$0.826 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

Original$3.301 (limited-time 60% off for off-peak usage) 20% off during peak hours

qwen3.7-plus-2026-05-26

Context cache enjoys a discount

Global

0<Token≤256K

$0.276

$1.101

$1.101

256K<Token≤1M

$0.826

$3.301

$3.301

qwen3.6-plus

Current capabilities are equivalent to qwen3.6-plus-2026-04-02

Context cache enjoys a discount

Global

0<Token≤256K

$0.5

$1.651

$1.651

256K<Token≤1M

$2

$6.602

$6.602

qwen3.6-plus-2026-04-02

Global

0<Token≤256K

$0.5

$1.651

$1.651

256K<Token≤1M

$2

$6.602

$6.602

Qwen Flash

Billing rule: Billing is based on input tokens and output tokens.

Factors that affect billing: If the model supports Batch calls, the unit prices of both input and output tokens are billed at 50% of the real-time inference price; if the model supports context caching, only input tokens enjoy a discount. The two cannot take effect at the same time.

NoteThe following models have no free quota in any region.

Singapore

Model IDService Deployment ScopePatternInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3.8-flash

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$0.15

$0.47

1 million Tokens

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Batch calls at half price

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤32K

$0.030

$0.130

1 million Tokens

32K<Token≤256K

$0.100

$0.400

256K<Token≤1M

$0.200

$0.800

qwen3.7-flash-2026-07-15

International

Non-thinking and thinking modes

0<Token≤32K

$0.030

$0.130

1 million Tokens

32K<Token≤256K

$0.100

$0.400

256K<Token≤1M

$0.200

$0.800

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Batch calls at half price

Context cache enjoys a discount

International

0<Token≤256K

$0.25

$1.5

1 million Tokens

256K<Token≤1M

$1

$4

qwen3.6-flash-2026-04-16

International

0<Token≤256K

$0.25

$1.5

1 million Tokens

256K<Token≤1M

$1

$4

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

Batch calls at half price

Context cache enjoys a discount

International

0<Token≤1M

$0.1

$0.4

1 million Tokens

qwen3.5-flash-2026-02-23

International

0<Token≤1M

$0.1

$0.4

1 million Tokens

qwen-flash

Current capabilities are equivalent to qwen-flash-2025-07-28

Batch calls at half price

Context cache enjoys a discount

International

0<Token≤256K

$0.05

$0.4

1 million Tokens

256K<Token≤1M

$0.25

$2

qwen-flash-2025-07-28

International

0<Token≤256K

$0.05

$0.4

1 million Tokens

256K<Token≤1M

$0.25

$2

China (Beijing)

Model IDInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.8-flash

Batch calls at half price

Context cache enjoys a discount

0<Token≤1M

$0.113

$0.382

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Batch calls at half price

Context cache enjoys a discount

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash-2026-07-15

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Batch calls at half price

Context cache enjoys a discount

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.6-flash-2026-04-16

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen3.5-flash-2026-02-23

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen-flash

Current capabilities are equivalent to qwen-flash-2025-07-28

Context cache enjoys a discount

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

qwen-flash-2025-07-28

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

Hong Kong (China)

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.8-flash

Context cache enjoys a discount

Hong Kong (China)

0<Token≤1M

$0.15

$0.47

qwen3.8-flash

Context cache enjoys a discount

Global

0<Token≤1M

$0.113

$0.382

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Context cache enjoys a discount

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash-2026-07-15

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

Context cache enjoys a discount

Hong Kong (China)

0<Token≤1M

$0.1

$0.4

qwen3.5-flash-2026-02-23

Hong Kong (China)

0<Token≤1M

$0.1

$0.4

Germany (Frankfurt)

Model IDService Deployment ScopeInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.8-flash

Context cache enjoys a discount

European Union

0<Token≤1M

$0.15

$0.47

qwen3.8-flash

Context cache enjoys a discount

Global

0<Token≤1M

$0.113

$0.382

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Context cache enjoys a discount

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash-2026-07-15

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash-2026-07-15

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

Global

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

Context cache enjoys a discount

European Union

0<Token≤1M

$0.1

$0.4

qwen3.5-flash-2026-02-23

Global

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen3.5-flash-2026-02-23

European Union

0<Token≤1M

$0.1

$0.4

qwen-flash

Current capabilities are equivalent to qwen-flash-2025-07-28

Context cache enjoys a discount

Global

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

qwen-flash-2025-07-28

Global

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

United States (Virginia)

Model IDService Deployment ScopeInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.8-flash

Context cache enjoys a discount

Global

0<Token≤1M

$0.113

$0.382

qwen3.8-flash

Context cache enjoys a discount

United States

0<Token≤1M

$0.15

$0.47

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Context cache enjoys a discount

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Context cache enjoys a discount

United States

0<Token≤32K

$0.030

$0.130

32K<Token≤256K

$0.100

$0.400

256K<Token≤1M

$0.200

$0.800

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.6-flash

United States

0<Token≤256K

$0.25

$1.5

256K<Token≤1M

$1

$4

qwen3.5-flash

Current capabilities are equivalent to qwen3.5-flash-2026-02-23

Global

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen3.5-flash-2026-02-23

Global

0<Token≤128K

$0.029

$0.287

128K<Token≤256K

$0.115

$1.147

256K<Token≤1M

$0.172

$1.72

qwen-flash

Current capabilities are equivalent to qwen-flash-2025-07-28

Context cache enjoys a discount

Global

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

qwen-flash

United States

0<Token≤256K

$0.05

$0.4

256K<Token≤1M

$0.25

$2

qwen-flash-2025-07-28

Global

0<Token≤128K

$0.022

$0.216

128K<Token≤256K

$0.087

$0.861

256K<Token≤1M

$0.173

$1.721

qwen-flash-2025-07-28

United States

0<Token≤256K

$0.05

$0.4

256K<Token≤1M

$0.25

$2

Japan (Tokyo)

Model IDService Deployment ScopeInput token range per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.8-flash

Context cache enjoys a discount

Global

0<Token≤1M

$0.113

$0.382

qwen3.7-flash

Current capabilities are equivalent to qwen3.7-flash-2026-07-15

Context cache enjoys a discount

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.7-flash-2026-07-15

Global

0<Token≤32K

$0.028

$0.110

32K<Token≤256K

$0.083

$0.330

256K<Token≤1M

$0.165

$0.660

qwen3.6-flash

Current capabilities are equivalent to qwen3.6-flash-2026-04-16

Context cache enjoys a discount

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

qwen3.6-flash-2026-04-16

Global

0<Token≤256K

$0.165

$0.99

256K<Token≤1M

$0.66

$3.961

Qwen Omni

Billing rule: Billing is based on input tokens and output tokens. For token calculation rules for different modalities, see Billing and rate limits.

NoteThe Qwen3.5-Omni and Qwen3-Omni series have a free quota only in the Singapore region, and no free quota in other regions.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashInternational$0.15$0.016$0.47
Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
Text/Image/VideoAudioText

Multimodal input

Text + Audio

Only audio is billed

qwen3.5-omni-plus

Its current capabilities are equivalent to qwen3.5-omni-plus-2026-03-15

International

$1.4

$11

$8.3

$44

1 million Tokens

qwen3.5-omni-plus-2026-03-15

International

$1.4

$11

$8.3

$44

None

qwen3.5-omni-flash

Its current capabilities are equivalent to qwen3.5-omni-flash-2026-03-15

International

$0.4

$3

$2.2

$11.9

None

qwen3.5-omni-flash-2026-03-15

International

$0.4

$3

$2.2

$11.9

None

More Models

Model IDService Deployment ScopePatternInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
TextAudioImage/VideoText

Text-only input

Text

Multimodal input

Text + Audio

Only audio is billed

qwen3-omni-flash

International

Non-thinking and thinking modes

$0.43

$3.81

$0.78

$1.66

$3.06

$15.11

None

China (Beijing)

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashthe Chinese mainland$0.113$0.014$0.382
Model IDInput unit price (per million tokens)Output unit price (per million tokens)
Text/Image/VideoAudioText

Multimodal input

Text + Audio

Only audio is billed

qwen3.5-omni-plus

Its current capabilities are equivalent to qwen3.5-omni-plus-2026-03-15

$0.96

$7.29

$5.5

$29.29

qwen3.5-omni-plus-2026-03-15

$0.96

$7.29

$5.5

$29.29

qwen3.5-omni-flash

Its current capabilities are equivalent to qwen3.5-omni-flash-2026-03-15

$0.3

$2.48

$1.83

$9.9

qwen3.5-omni-flash-2026-03-15

$0.3

$2.48

$1.83

$9.9

More Models

Model IDService Deployment ScopePatternInput unit price (per million tokens)Output unit price (per million tokens)
TextAudioImage/VideoText

Text-only input

Text

Multimodal input

Text + Audio

Only audio is billed

qwen3-omni-flash

the Chinese mainland

Non-thinking and thinking modes

$0.258

$2.265

$0.473

$0.989

$1.821

$8.974

Hong Kong (China)

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashGlobal$0.113$0.014$0.382

Japan (Tokyo)

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashGlobal$0.113$0.014$0.382

Germany (Frankfurt)

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashGlobal$0.113$0.014$0.382

United States (Virginia)

Model IDService Deployment ScopeInput unit price (per million tokens)Cache-hit input unit price (per million tokens)Output unit price (per million tokens)
qwen3.8-omni-flashGlobal$0.113$0.014$0.382

Qwen Omni-Realtime

Billing rule: You are charged for both input and output tokens. For token calculation rules for different modalities, see Billing and rate limits.

When outputting speech, the audio and corresponding text of qwen3.8-omni-flash-realtime are billed separately at the audio output and text output unit prices; the Qwen3.5-Omni-Realtime series only charges for audio, and the corresponding text is not billed.

NoteThe free quota indicated in the table applies only to the Singapore region; other regions have no free quota.

Singapore

Unit price: USD/million tokens.

Model ID

Input: Text/Image/Video

Input: Audio

Output: Text

Output: Audio

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3.8-omni-flash-realtime

$0.23

$0.93

$0.70

$1.87

1 million Tokens

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
Text/ImageAudioText

Multimodal input

Text + Audio

Only audio is billed

qwen3.5-omni-plus-realtime

Its current capabilities are equivalent to qwen3.5-omni-plus-realtime-2026-03-15

International

$2.1

$16.5

$12.4

$62

1 million Tokens

qwen3.5-omni-plus-realtime-2026-03-15

International

$2.1

$16.5

$12.4

$62

None

qwen3.5-omni-flash-realtime

Its current capabilities are equivalent to qwen3.5-omni-flash-realtime-2026-03-15

International

$0.55

$4.5

$3.3

$17.7

None

qwen3.5-omni-flash-realtime-2026-03-15

International

$0.55

$4.5

$3.3

$17.7

None

More Models

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
TextAudioImageText

Text-only input

Text

Multimodal input

Text + Audio

Only audio is billed

qwen3-omni-flash-realtime-2025-09-15

International

$0.52

$4.57

$0.94

$1.99

$3.67

$18.13

None

China (Beijing)

Unit price: USD/million tokens.

Model ID

Input: Text/Image/Video

Input: Audio

Output: Text

Output: Audio

qwen3.8-omni-flash-realtime

$0.212

$0.848

$0.636

$1.696

Model IDInput unit price (per million tokens)Output unit price (per million tokens)
Text/ImageAudioText

Multimodal input

Text + Audio

Only audio is billed

qwen3.5-omni-plus-realtime

Its current capabilities are equivalent to qwen3.5-omni-plus-realtime-2026-03-15

$1.38

$11

$8.25

$41.26

qwen3.5-omni-plus-realtime-2026-03-15

$1.38

$11

$8.25

$41.26

qwen3.5-omni-flash-realtime

Its current capabilities are equivalent to qwen3.5-omni-flash-realtime-2026-03-15

$0.45

$3.71

$2.75

$14.71

qwen3.5-omni-flash-realtime-2026-03-15

$0.45

$3.71

$2.75

$14.71

More Models

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)
TextAudioImageText

Text-only input

Text

Multimodal input

Text + Audio

Only audio is billed

qwen3-omni-flash-realtime-2025-09-15

the Chinese mainland

$0.315

$2.709

$0.559

$1.19

$2.179

$10.766

Qwen VL

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-vl-plus

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤32K

$0.2

$1.6

1 million Tokens

32K<Token≤128K

$0.3

$2.4

128K<Token≤256K

$0.6

$4.8

China (Beijing)

Model IDPatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen3-vl-plus

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤32K

$0.143

$1.434

32K<Token≤128K

$0.215

$2.15

128K<Token≤256K

$0.43

$4.301

Hong Kong (China)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen3-vl-plus

Context cache enjoys a discount

Hong Kong (China)

Non-thinking and thinking modes

0<Token≤32K

$0.2

$1.6

32K<Token≤128K

$0.3

$2.4

128K<Token≤256K

$0.6

$4.8

Germany (Frankfurt)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen3-vl-plus

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤32K

$0.143

$1.434

32K<Token≤128K

$0.215

$2.15

128K<Token≤256K

$0.43

$4.301

qwen3-vl-plus

Context cache enjoys a discount

European Union

Non-thinking and thinking modes

0<Token≤32K

$0.2

$1.6

32K<Token≤128K

$0.3

$2.4

128K<Token≤256K

$0.6

$4.8

United States (Virginia)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen3-vl-plus

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤32K

$0.143

$1.434

32K<Token≤128K

$0.215

$2.15

128K<Token≤256K

$0.43

$4.301

Qwen OCR

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen-vl-ocr

Current capabilities are equivalent to qwen-vl-ocr-2025-11-20

International

$0.07

$0.16

1 million Tokens

qwen-vl-ocr-2025-11-20

International

$0.07

$0.16

None

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)

qwen3.5-ocr

$0.069

$0.275

qwen-vl-ocr

Current capabilities are equivalent to qwen-vl-ocr-2025-11-20

$0.043

$0.072

qwen-vl-ocr-2025-11-20

$0.043

$0.072

qwen-vl-ocr-2024-10-28

$0.717

$0.717

Germany (Frankfurt)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-vl-ocr

Current capabilities are equivalent to qwen-vl-ocr-2025-11-20

Global

$0.043

$0.072

qwen-vl-ocr-2025-11-20

Global

$0.043

$0.072

United States (Virginia)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-vl-ocr

Current capabilities are equivalent to qwen-vl-ocr-2025-11-20

Global

$0.043

$0.072

qwen-vl-ocr-2025-11-20

Global

$0.043

$0.072

Qwen Math models

Billing rule: Billing is based on input tokens and output tokens.

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

qwen-math-plus-2024-09-19

$0.574

$1.721

None

Qwen Coder

Billing rule: Billing is based on input tokens and output tokens.

Billing factors: If the model supports context cache, only input tokens are eligible for the discount.

NoteThe following models have no free quota in any region.

Singapore

Model IDService Deployment ScopeInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-coder-flash

Current capabilities are equivalent to qwen3-coder-flash-2025-07-28

International

0<Token≤32K

$0.3

$1.5

None

32K<Token≤128K

$0.5

$2.5

128K<Token≤256K

$0.8

$4

256K<Token≤1M

$1.6

$9.6

qwen3-coder-flash-2025-07-28

International

0<Token≤32K

$0.3

$1.5

None

32K<Token≤128K

$0.5

$2.5

128K<Token≤256K

$0.8

$4

256K<Token≤1M

$1.6

$9.6

China (Beijing)

Qwen3-Coder series models
Model IDInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3-coder-flash

Current capabilities are equivalent to qwen3-coder-flash-2025-07-28

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

qwen3-coder-flash-2025-07-28

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

Germany (Frankfurt)

Model IDService Deployment ScopeInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3-coder-flash

Current capabilities are equivalent to qwen3-coder-flash-2025-07-28

Context cache enjoys a discount

Global

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

qwen3-coder-flash-2025-07-28

Global

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

United States (Virginia)

Model IDService Deployment ScopeInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)

qwen3-coder-flash

Current capabilities are equivalent to qwen3-coder-flash-2025-07-28

Context cache enjoys a discount

Global

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

qwen3-coder-flash-2025-07-28

Global

0<Token≤32K

$0.144

$0.574

32K<Token≤128K

$0.216

$0.861

128K<Token≤256K

$0.359

$1.434

256K<Token≤1M

$0.717

$3.584

Qwen translation models

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-mt-plus

International

$2.46

$7.37

1 million Tokens

qwen-mt-flash

International

$0.16

$0.49

1 million Tokens

qwen-mt-lite

International

$0.12

$0.36

1 million Tokens

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-mt-plus

$0.259

$0.775

qwen-mt-flash

$0.101

$0.280

qwen-mt-lite

$0.086

$0.229

Germany (Frankfurt)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-mt-plus

Global

$0.259

$0.775

qwen-mt-flash

Global

$0.101

$0.280

qwen-mt-lite

Global

$0.086

$0.229

United States (Virginia)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-mt-flash

Global

$0.101

$0.280

qwen-mt-lite

Global

$0.086

$0.229

qwen-mt-lite

United States

$0.12

$0.36

qwen-mt-plus

Global

$0.259

$0.775

Qwen data mining models

Billing rule: Billing is based on input tokens and output tokens.

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

qwen-doc-turbo

$0.087

$0.144

None

Qwen deep research models

Billing rule: Billing is based on input tokens and output tokens.

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

qwen-deep-research

$7.742

$23.367

None

Text generation - Qwen open-source version

Qwen3.8

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3.8-2.4t-a95b

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$2

$6

1 million Tokens

qwen3.8-27b

Context cache enjoys a discount

International

Non-thinking and thinking modes

0<Token≤1M

$0.5

$3

1 million Tokens

China (Beijing)

Model IDPatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought + response

qwen3.8-2.4t-a95b

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$1.650

$4.951

qwen3.8-27b

Context cache enjoys a discount

Non-thinking and thinking modes

0<Token≤1M

$0.424

$1.696

Qwen3.6

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.6-35b-a3b

International

0<Token≤256K

$0.375

$2.25

$2.25

None

qwen3.6-27b

International

0<Token≤256K

$0.6

$3.6

$3.6

None

China (Beijing)

Model ID

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.6-35b-a3b

0<Token≤256K

$0.248

$1.485

$1.485

qwen3.6-27b

0<Token≤256K

$0.412564

$2.475384

$2.475384

Germany (Frankfurt)

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.6-35b-a3b

Global

0<Token≤256K

$0.248

$1.485

$1.485

United States (Virginia)

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.6-35b-a3b

Global

0<Token≤256K

$0.248

$1.485

$1.485

Qwen3.5

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.5-397b-a17b

International

0<Token≤256K

$0.6

$3.6

$3.6

None

qwen3.5-122b-a10b

International

0<Token≤256K

$0.4

$3.2

$3.2

None

qwen3.5-27b

International

0<Token≤256K

$0.3

$2.4

$2.4

None

qwen3.5-35b-a3b

International

0<Token≤256K

$0.25

$2

$2

None

China (Beijing)

Model ID

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.5-397b-a17b

0<Token≤128K

$0.172

$1.032

$1.032

128K<Token≤256K

$0.43

$2.58

$2.58

qwen3.5-122b-a10b

0<Token≤128K

$0.115

$0.917

$0.917

128K<Token≤256K

$0.287

$2.294

$2.294

qwen3.5-27b

0<Token≤128K

$0.086

$0.688

$0.688

128K<Token≤256K

$0.258

$2.064

$2.064

qwen3.5-35b-a3b

0<Token≤128K

$0.057

$0.459

$0.459

128K<Token≤256K

$0.229

$1.835

$1.835

Germany (Frankfurt)

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.5-397b-a17b

Global

0<Token≤128K

$0.172

$1.032

$1.032

128K<Token≤256K

$0.43

$2.58

$2.58

qwen3.5-122b-a10b

Global

0<Token≤128K

$0.115

$0.917

$0.917

128K<Token≤256K

$0.287

$2.294

$2.294

qwen3.5-27b

Global

0<Token≤128K

$0.086

$0.688

$0.688

128K<Token≤256K

$0.258

$2.064

$2.064

qwen3.5-35b-a3b

Global

0<Token≤128K

$0.057

$0.459

$0.459

128K<Token≤256K

$0.229

$1.835

$1.835

United States (Virginia)

Model ID

Service Deployment Scope

Input token range per request

Input unit price (per million tokens)

Output unit price (per million tokens)

Non-thinking mode

Thinking mode (chain-of-thought + response)

qwen3.5-397b-a17b

Global

0<Token≤128K

$0.172

$1.032

$1.032

128K<Token≤256K

$0.43

$2.58

$2.58

qwen3.5-122b-a10b

Global

0<Token≤128K

$0.115

$0.917

$0.917

128K<Token≤256K

$0.287

$2.294

$2.294

qwen3.5-27b

Global

0<Token≤128K

$0.086

$0.688

$0.688

128K<Token≤256K

$0.258

$2.064

$2.064

qwen3.5-35b-a3b

Global

0<Token≤128K

$0.057

$0.459

$0.459

128K<Token≤256K

$0.229

$1.835

$1.835

Qwen-VL

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models have no free quota in any region.

China (Beijing)

More Models

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen2-vl-72b-instruct

the Chinese mainland

$2.294

$6.881

qwen2-vl-7b-instruct

the Chinese mainland

Limited Time Free

qwen2-vl-2b-instruct

the Chinese mainland

Limited Time Free

Text generation - third-party models

DeepSeek

Billing rule: Billing is based on input tokens and output tokens.

Some models are billed at peak and off-peak rates (see the peak and off-peak unit prices marked in the table below): Peak hours are from 08:00 to 22:00 (UTC+8), and off-peak hours are from 22:00 to 08:00 the next day (UTC+8).

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

NoteDeepSeek-V4-Flash-0731 switched to the peak/off-peak pricing model at 2026-08-17 00:00. The unit prices after the adjustment are shown in the table below. For more information, see the DeepSeek-V4-Flash-0731 model price increase notice.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

International

Peak $0.3

Off-peak $0.15

Peak $1.2

Off-peak $0.6

None

deepseek-v4.1-flash

Context cache enjoys a discount

International

Peak $0.3

Off-peak $0.15

Peak $1.2

Off-peak $0.6

1 million Tokens

deepseek-v4-pro

Context cache enjoys a discount

International

$2.400

$4.800

None

deepseek-v4-pro-0813

Context cache enjoys a discount

International

Peak $1.32

Off-peak $0.66

Peak $3.96

Off-peak $1.98

1 million Tokens

deepseek-v4-flash-0731

Context cache enjoys a discount

International

Peak $0.44

Off-peak $0.22

Peak $1.32

Off-peak $0.66

1 million Tokens

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

No free quota

deepseek-v4.1-flash

Context cache enjoys a discount

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

No free quota

deepseek-v4-pro

Context cache enjoys a discount

$1.65

$3.301

No free quota

deepseek-v4-pro-0813

Context cache enjoys a discount

Peak $1.272

Off-peak $0.636

Peak $3.816

Off-peak $1.908

No free quota

deepseek-v4-flash-0731

Context cache enjoys a discount

Peak $0.424

Off-peak $0.212

Peak $1.272

Off-peak $0.636

No free quota

deepseek-r1-distill-qwen-1.5b

Limited Time Free

deepseek-r1-distill-llama-70b

Limited Time Free

Germany (Frankfurt)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4.1-flash

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4-pro

Context cache enjoys a discount

Global

$1.65

$3.301

deepseek-v4-pro-0813

Context cache enjoys a discount

Global

Peak $1.272

Off-peak $0.636

Peak $3.816

Off-peak $1.908

deepseek-v4-flash-0731

Context cache enjoys a discount

Global

Peak $0.424

Off-peak $0.212

Peak $1.272

Off-peak $0.636

United States (Virginia)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4.1-flash

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4-pro

Context cache enjoys a discount

Global

$1.65

$3.301

deepseek-v4-pro-0813

Context cache enjoys a discount

Global

Peak $1.272

Off-peak $0.636

Peak $3.816

Off-peak $1.908

deepseek-v4-pro

Context cache enjoys a discount

International

$2.400

$4.800

deepseek-v4-flash-0731

Context cache enjoys a discount

Global

Peak $0.424

Off-peak $0.212

Peak $1.272

Off-peak $0.636

deepseek-v4-flash-0731

Context cache enjoys a discount

International

Peak $0.424

Off-peak $0.212

Peak $1.32

Off-peak $0.66

Japan (Tokyo)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4.1-flash

Context cache enjoys a discount

Global

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4-pro

Context cache enjoys a discount

Global

$1.65

$3.301

deepseek-v4-pro-0813

Context cache enjoys a discount

Global

Peak $1.272

Off-peak $0.636

Peak $3.816

Off-peak $1.908

deepseek-v4-flash-0731

Context cache enjoys a discount

Global

Peak $0.424

Off-peak $0.212

Peak $1.272

Off-peak $0.636

deepseek-v4-pro

Context cache enjoys a discount

Japan

$2.4

$4.8

deepseek-v4-flash-0731

Context cache enjoys a discount

Japan

Peak $0.44

Off-peak $0.22

Peak $1.32

Off-peak $0.66

Hong Kong (China)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

deepseek-v4.1-flash-1017

Context cache enjoys a discount

Hong Kong (China)

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4.1-flash

Context cache enjoys a discount

Hong Kong (China)

Peak $0.283

Off-peak $0.141

Peak $1.131

Off-peak $0.565

deepseek-v4-flash-0731

Context cache enjoys a discount

Hong Kong (China)

Peak $0.424

Off-peak $0.212

Peak $1.272

Off-peak $0.636

deepseek-v4-pro-0813

Context cache enjoys a discount

Hong Kong (China)

Peak $1.272

Off-peak $0.636

Peak $3.816

Off-peak $1.908

Kimi

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models have no free quota in any region.

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

kimi-k3.1-pro

$1.837

Cache hit input: $0.184

$9.611

None

kimi-k3

$2.827

$14.133

None

kimi-k2.7-code

$0.894

$3.713

None

kimi-k2.6

$0.8939

$3.7131

None

kimi-k2.5

$0.574

$3.011

None

Germany (Frankfurt)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

kimi-k3.1-pro

Global

$1.837

Cache hit input: $0.184

$9.611

kimi-k3

Global

$2.827

$14.133

kimi-k2.7-code

Global

$0.894

$3.713

kimi-k2.5

Global

$0.574

$3.011

United States (Virginia)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

kimi-k3.1-pro

Global

$1.837

Cache hit input: $0.184

$9.611

kimi-k3

Global

$2.827

$14.133

kimi-k3.1-pro

United States

$2

Cache hit input: $0.2

$10

kimi-k3

International

$3

$15

kimi-k2.7-code

Global

$0.894

$3.713

kimi-k2.5

Global

$0.574

$3.011

Hong Kong (China)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

kimi-k3.1-pro

Global

$1.837

Cache hit input: $0.184

$9.611

kimi-k3

Global

$2.827

$14.133

kimi-k2.7-code

Global

$0.894

$3.713

Japan (Tokyo)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

kimi-k3.1-pro

Global

$1.837

Cache hit input: $0.184

$9.611

kimi-k3

Context cache offers a discount

Global

$2.827

$14.133

kimi-k2.7-code

Context cache offers a discount

Global

$0.894

$3.713

kimi-k2.5

Context cache offers a discount

Global

$0.574

$3.011

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

kimi-k3.1-pro

International

$2

Cache hit input: $0.2

$10

kimi-k3

International

$3

$15

kimi-k2.7-code

International

$0.95

$4

MiniMax

Billing rule: Billing is based on input tokens and output tokens.

China (Beijing)

Model IDPatternInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

MiniMax-M2.5

Thinking mode only

$0.304

$1.213

GLM

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

glm-5.3

International

Non-thinking and thinking modes

No tier distinction

$1.400

$4.400

1 million Tokens

glm-5.2

International

Non-thinking and thinking modes

No tier distinction

$1.400

$4.400

None

glm-5.2-fast-preview

International

Non-thinking and thinking modes

No tier distinction

$2.800

$8.800

None

glm-5.3-prime

International

Non-thinking and thinking modes

No tier distinction

$2.800

$8.800

None

glm-5.1

International

Non-thinking and thinking modes

0<Token≤200K

$1.400

$4.400

None

China (Beijing)

Model IDPatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

glm-5.3

Non-thinking and thinking modes

No tier distinction

$1.131

$3.957

glm-5.2

Non-thinking and thinking modes

No tier distinction

$1.100

$3.851

glm-5.2-fast-preview

Non-thinking and thinking modes

No tier distinction

$2.200

$7.702

glm-5.1

Non-thinking and thinking modes

0<Token≤32K

$0.825

$3.301

32K<Token≤200K

$1.100

$3.851

glm-5

Non-thinking and thinking modes

0<Token≤32K

$0.573

$2.580

32K<Token≤166K

$0.860

$3.154

Germany (Frankfurt)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

glm-5.3

Global

Non-thinking and thinking modes

No tier distinction

$1.131

$3.957

glm-5.2

Global

Non-thinking and thinking modes

No tier distinction

$1.1

$3.851

glm-5.1

Global

Non-thinking and thinking modes

0<Token≤32K

$0.825

$3.301

32K<Token≤200K

$1.100

$3.851

United States (Virginia)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

glm-5.3

Global

Non-thinking and thinking modes

No tier distinction

$1.131

$3.957

glm-5.3

United States

Non-thinking and thinking modes

No tier distinction

$1.4

$4.4

glm-5.2

Global

Non-thinking and thinking modes

No tier distinction

$1.1

$3.851

glm-5.2

International

Non-thinking and thinking modes

No tier distinction

$1.4

$4.4

glm-5.1

Global

Non-thinking and thinking modes

0<Token≤32K

$0.825

$3.301

32K<Token≤200K

$1.100

$3.851

Hong Kong (China)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

glm-5.3

Global

Non-thinking and thinking modes

No tier distinction

$1.131

$3.957

glm-5.2

Global

Non-thinking and thinking modes

No tier distinction

$1.1

$3.851

Japan (Tokyo)

Model IDService Deployment ScopePatternInput tokens per requestInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

glm-5.3

Global

Non-thinking and thinking modes

No tier distinction

$1.131

$3.957

glm-5.1

Context cache enjoys a discount

Global

Non-thinking and thinking modes

0<Token≤32K

$0.825

$3.301

32K<Token≤200K

$1.100

$3.851

GLM-Zhipu

Billing rule: Billing is based on input tokens and output tokens.

Singapore

Model IDService Deployment ScopePatternInput unit price (per million tokens)Output unit price (per million tokens)
Chain-of-thought and response

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

ZHIPU/GLM-5.3

International

Non-thinking and thinking modes

$1.400

$4.400

None

ZHIPU/GLM-5.2

International

Non-thinking and thinking modes

$1.400

$4.400

None

MindLab

Billing rule: Fees are based on input tokens and output tokens. Input tokens that hit the cache and tokens used to create explicit caches are billed at separate unit prices. For more information, see the model information page and Context cache.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

mindlab/macaron-v1-tall

International

$0.45

$2.60

None

mindlab/macaron-v1.1-venti

International

$1.19

$3.74

None

mindlab/macaron-v1.1-coding-venti

International

$1.19

$3.74

None

Image Generation

Billing rule: Fees are based on input images and the number of images successfully generated. For models that do not specify an input image price, input is not billed; only output is billed.

Billing formula: Fee = Input image unit price × Number of input images + Output image unit price × Number of output images.

Billing details: Failed requests do not incur any fees or consume the free quota.

Billing example: Some images fail to be generated

Assume the input image unit price is 0 and the output image unit price is $0.10/image. If you call the API to request 4 images, but only 3 image URLs are successfully returned and 1 image fails to be generated, the system bills only the successfully generated images.

  • Billed quantity: 3 images.
  • Fee calculation: 0.1 × 3 = $0.3.

Qwen Image Generation and Editing

Fees are based on the number of input and output images. For billing rules, see Image Generation.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output image resolution

Input unit price

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-image-3.0-pro

International

1k

$0.003/image

$0.04/image

10 images in total for input and output

2k

$0.075/image

qwen-image-3.0

International

1k

$0.003/image

$0.03/image

10 images in total for input and output

2k

$0.03/image

China (Beijing)

Model ID

Output image resolution

Input unit price

Output unit price

qwen-image-3.0-pro

1k

$0.002750/image

$0.034380/image

2k

$0.068761/image

qwen-image-3.0

1k

$0.002750/image

$0.024754/image

2k

$0.024754/image

Hong Kong (China)

Model ID

Service Deployment Scope

Output image resolution

Input unit price

Output unit price

qwen-image-3.0-pro

Global

1k

$0.002750/image

$0.034380/image

2k

$0.068761/image

qwen-image-3.0

Global

1k

$0.002750/image

$0.024754/image

2k

$0.024754/image

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output image resolution

Input unit price

Output unit price

qwen-image-3.0-pro

Global

1k

$0.002750/image

$0.034380/image

2k

$0.068761/image

qwen-image-3.0

Global

1k

$0.002750/image

$0.024754/image

2k

$0.024754/image

United States (Virginia)

Model ID

Service Deployment Scope

Output image resolution

Input unit price

Output unit price

qwen-image-3.0-pro

Global

1k

$0.002750/image

$0.034380/image

2k

$0.068761/image

qwen-image-3.0

Global

1k

$0.002750/image

$0.024754/image

2k

$0.024754/image

Japan (Tokyo)

Model ID

Service Deployment Scope

Output image resolution

Input unit price

Output unit price

qwen-image-3.0-pro

Global

1k

$0.002750/image

$0.034380/image

2k

$0.068761/image

qwen-image-3.0

Global

1k

$0.002750/image

$0.024754/image

2k

$0.024754/image

Qwen Image Generation and Editing 2.0 series

The qwen-image-2.0 series is billed by image and supports text-to-image and image-to-image/image editing. Unit prices by region are as follows.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota

qwen-image-2.1-pro

International

$0.04/image

10 images

qwen-image-2.1-turbo

International

$0.016/image

10 images

qwen-image-2.0-pro

International

$0.075/image

None

qwen-image-2.0-pro-2026-06-22

International

$0.075/image

None

qwen-image-2.0-pro-2026-04-22

International

$0.075/image

None

qwen-image-2.0-pro-2026-03-03

International

$0.075/image

None

qwen-image-2.0

International

$0.035/image

None

qwen-image-2.0-2026-03-03

International

$0.035/image

None

China (Beijing)

Model ID

Service Deployment Scope

Output unit price

qwen-image-2.1-pro

Global

$0.035333/image

qwen-image-2.1-turbo

Global

$0.014133/image

qwen-image-2.0-pro

Global

$0.071676/image

qwen-image-2.0-pro-2026-06-22

Global

$0.071676/image

qwen-image-2.0-pro-2026-04-22

Global

$0.071676/image

qwen-image-2.0-pro-2026-03-03

Global

$0.071676/image

qwen-image-2.0

Global

$0.028671/image

qwen-image-2.0-2026-03-03

Global

$0.028671/image

Hong Kong (China)

Model ID

Service Deployment Scope

Output unit price

qwen-image-2.1-pro

Global

$0.035333/image

qwen-image-2.1-turbo

Global

$0.014133/image

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output unit price

qwen-image-2.1-pro

Global

$0.035333/image

qwen-image-2.1-turbo

Global

$0.014133/image

United States (Virginia)

Model ID

Service Deployment Scope

Output unit price

qwen-image-2.1-pro

Global

$0.035333/image

qwen-image-2.1-turbo

Global

$0.014133/image

Japan (Tokyo)

Model ID

Service Deployment Scope

Output unit price

qwen-image-2.1-pro

Global

$0.035333/image

qwen-image-2.1-turbo

Global

$0.014133/image

Qwen Image Translation

Billing only. For billing rules, see Image Generation.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota(Note)

qwen-mt-image-2.0

International

$0.0006/image

100 images

China (Beijing)

Model ID

Output unit price

Free quota(Note)

qwen-mt-image-2.0

$0.000565/image

None

qwen-mt-image

$0.000431/image

None

Qwen-Text-to-Image-Z-Image

Billing only. For billing rules, see Image Generation.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

z-image-turbo

International

With prompt rewriting disabled (prompt_extend=false): $0.015/image

With prompt rewriting enabled (prompt_extend=true): $0.03/image

100 images

China (Beijing)

Model ID

Output unit price

z-image-turbo

With prompt rewriting disabled (prompt_extend=false): $0.01434/image

With prompt rewriting enabled (prompt_extend=true): $0.02868/image

Wanx text-to-image

Billing only. For billing rules, see Image Generation.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.6-t2i

International

$0.03/image

None

wan2.5-t2i-preview

International

$0.03/image

None

wan2.2-t2i-plus

International

$0.05/image

None

wan2.2-t2i-flash

International

$0.025/image

None

wan2.1-t2i-plus

International

$0.05/image

200 images

wan2.1-t2i-turbo

International

$0.025/image

200 images

China (Beijing)

Model ID

Output unit price

wan2.6-t2i

$0.028671/image

wan2.5-t2i-preview

$0.028671/image

wan2.2-t2i-plus

$0.020070/image

wan2.2-t2i-flash

$0.028671/image

wanx2.1-t2i-plus

$0.028671/image

wanx2.1-t2i-turbo

$0.020070/image

wanx2.0-t2i-turbo

$0.005735/image

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output unit price

wan2.6-t2i

Global

$0.028671/image

United States (Virginia)

Model ID

Service Deployment Scope

Output unit price

wan2.6-t2i

Global

$0.028671/image

Wanx Image Generation and Editing

Billing only. For billing rules, see Image Generation.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.7-image-pro

International

$0.075/image

50 images

wan2.7-image

International

$0.03/image

50 images

wan2.6-image

International

$0.03/image

None

China (Beijing)

Model ID

Output unit price

wan2.7-image-pro

$0.068761/image

wan2.7-image

$0.027504/image

wan2.6-image

$0.028671/image

United States (Virginia)

Model ID

Service Deployment Scope

Output unit price

wan2.6-image

Global

$0.028671/image

Wanx General Image Editing

Billing only. For billing rules, see Image Generation.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.5-i2i-preview

International

$0.03/image

None

China (Beijing)

Model ID

Output unit price

wan2.5-i2i-preview

$0.028671/image

wanx2.1-imageedit

$0.020070/image

Vidu-Image Generation

Singapore

Model ID

Output image resolution

Unit price

Free quota(Note)

vidu/vidu-image_reference2image

1K

$0.1/image

None

2K

$0.16/image

4K

$0.24/image

Video Generation

HappyHorse-T2V

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Different output video resolutions (480P/720P/1080P) have different unit prices; failed requests do not incur fees or consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

happyhorse-1.1-t2v

International

480P

$0.07/s

10 s

720P

$0.14/s

1080P

$0.18/s

happyhorse-1.0-t2v

International

720P

$0.14/s

None

1080P

$0.24/s

China (Beijing)

Model ID

Output video resolution

Output unit price

happyhorse-1.1-t2v

480P

$0.061885/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-t2v

720P

$0.123769/s

1080P

$0.220034/s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-t2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-t2v

Global

720P

$0.123769/s

1080P

$0.220034/s

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-t2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-t2v

Global

720P

$0.123769/s

1080P

$0.220034/s

Japan (Tokyo)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-t2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

HappyHorse-I2V-First Frame

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Different output video resolutions (480P/720P/1080P) have different unit prices; failed requests do not incur fees or consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

happyhorse-1.1-i2v

International

480P

$0.07/s

10 s

720P

$0.14/s

1080P

$0.18/s

happyhorse-1.0-i2v

International

720P

$0.14/s

None

1080P

$0.24/s

China (Beijing)

Model ID

Output video resolution

Output unit price

happyhorse-1.1-i2v

480P

$0.061885/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-i2v

720P

$0.123769/s

1080P

$0.220034/s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-i2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-i2v

Global

720P

$0.123769/s

1080P

$0.220034/s

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-i2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-i2v

Global

720P

$0.123769/s

1080P

$0.220034/s

Japan (Tokyo)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-i2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

HappyHorse-reference-to-video

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Different output video resolutions (480P/720P/1080P) have different unit prices; failed requests do not incur fees or consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

happyhorse-1.1-r2v

International

480P

$0.07/s

10 s

720P

$0.14/s

1080P

$0.18/s

happyhorse-1.0-r2v

International

720P

$0.14/s

None

1080P

$0.24/s

China (Beijing)

Model ID

Output video resolution

Output unit price

happyhorse-1.1-r2v

480P

$0.061885/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-r2v

720P

$0.123769/s

1080P

$0.220034/s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-r2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-r2v

Global

720P

$0.123769/s

1080P

$0.220034/s

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-r2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

happyhorse-1.0-r2v

Global

720P

$0.123769/s

1080P

$0.220034/s

Japan (Tokyo)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

happyhorse-1.1-r2v

Global

480P

$0.0618845/s

720P

$0.123769/s

1080P

$0.165026/s

HappyHorse-Video Editing

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Billing rule: Both input videos and output videos are billed by video seconds.

Failed requests do not incur fees or consume the free quota.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

happyhorse-1.0-video-edit

International

720P

$0.14/s

10 s

1080P

$0.24/s

China (Beijing)

Model ID

Output video resolution

Input and output unit prices

happyhorse-1.0-video-edit

720P

$0.123769/s

1080P

$0.220034/s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

happyhorse-1.0-video-edit

Global

720P

$0.123769/s

1080P

$0.220034/s

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

happyhorse-1.0-video-edit

Global

720P

$0.123769/s

1080P

$0.220034/s

Japan (Tokyo)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

happyhorse-1.0-video-edit

Global

720P

$0.123769/s

1080P

$0.220034/s

Wan3.0-Video Generation

Billing rule: Both input videos and output videos are billed by video seconds.

Failed requests do not incur fees or consume the free quota.

Billing formula: Billable duration = Input video duration + Output video duration.

  • The billable duration of an input video is the actual number of video seconds input.
  • The billable duration of an output video is the number of video seconds successfully generated.
  • The free quota is a total of 30 seconds for input video duration and output video duration combined.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan3.0-video-prime

International

480P

$0.068/ s

None

720P

$0.14/s

1080P

$0.28/ s

wan3.0-video

International

480P

Original$0.05/ s (30% off for a limited time)

30 s

720P

Original$0.1/ s (30% off for a limited time)

1080P

Original$0.2/ s (30% off for a limited time)

China (Beijing)

Model ID

Output video resolution

Input and output unit prices

wan3.0-video-prime

480P

$0.0636/ s

720P

$0.127199/ s

1080P

$0.254399/ s

wan3.0-video

480P

Original$0.041256/ s (30% off for a limited time)

720P

Original$0.082513/ s (30% off for a limited time)

1080P

Original$0.165025/ s (30% off for a limited time)

Japan (Tokyo)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

wan3.0-video-prime

Global

480P

$0.0636/ s

720P

$0.127199/ s

1080P

$0.254399/ s

wan3.0-video

Global

480P

Original$0.041256/ s (30% off for a limited time)

720P

Original$0.082513/ s (30% off for a limited time)

1080P

Original$0.165025/ s (30% off for a limited time)

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

wan3.0-video-prime

Global

480P

$0.0636/ s

720P

$0.127199/ s

1080P

$0.254399/ s

wan3.0-video

Global

480P

Original$0.041256/ s (30% off for a limited time)

720P

Original$0.082513/ s (30% off for a limited time)

1080P

Original$0.165025/ s (30% off for a limited time)

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

wan3.0-video-prime

Global

480P

$0.0636/ s

720P

$0.127199/ s

1080P

$0.254399/ s

wan3.0-video

Global

480P

Original$0.041256/ s (30% off for a limited time)

720P

Original$0.082513/ s (30% off for a limited time)

1080P

Original$0.165025/ s (30% off for a limited time)

Hong Kong (China)

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

wan3.0-video-prime

Global

480P

$0.0636/ s

720P

$0.127199/ s

1080P

$0.254399/ s

wan3.0-video

Global

480P

Original$0.041256/ s (30% off for a limited time)

720P

Original$0.082513/ s (30% off for a limited time)

1080P

Original$0.165025/ s (30% off for a limited time)

Wan-T2V

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Different output video resolutions (480P/720P/1080P) have different unit prices; failed requests do not incur fees or consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.7-t2v-2026-06-12

International

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.7-t2v-2026-04-25

International

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.7-t2v

International

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.6-t2v

International

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.5-t2v-preview

International

480P

$0.05/ s

None

720P

$0.10/ s

1080P

$0.15/ s

wan2.2-t2v-plus

International

480P

$0.02/ s

None

1080P

$0.10/ s

wan2.1-t2v-turbo

International

480P

$0.036/ s

50 s

720P

$0.036/ s

wan2.1-t2v-plus

International

720P

$0.10/ s

50 s

China (Beijing)

Model ID

Output video resolution

Output unit price

wan2.7-t2v-2026-06-12

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.7-t2v-2026-04-25

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.7-t2v

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.6-t2v

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.5-t2v-preview

480P

$0.043006/ s

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.2-t2v-plus

480P

$0.02007/ s

1080P

$0.100347/ s

wanx2.1-t2v-turbo

480P

$0.034405/ s

720P

$0.034405/ s

wanx2.1-t2v-plus

720P

$0.100347/ s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

wan2.6-t2v

Global

720P

$0.086012/ s

1080P

$0.143353/ s

United States (Virginia)

Model ID

Service Deployment Scope

Output video resolution

Output unit price

wan2.6-t2v

Global

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.6-t2v

United States

720P

$0.1/ s

1080P

$0.15/ s

Wan-I2V

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution; failed requests incur no fees and do not consume the free quota.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Output video type

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.7-i2v-2026-04-25

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.7-i2v

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

China (Beijing)

Model ID

Output video type

Output video resolution

Output unit price

wan2.7-i2v-2026-04-25

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.7-i2v

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

Wan-I2V based on first frame

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution and video type (with audio/silent); failed requests incur no fees and do not consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video type

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.6-i2v-flash

International

Video with audio

audio=true

720P

$0.05/ s

None

1080P

$0.075/ s

Silent video

audio=false

720P

$0.025/ s

1080P

$0.0375/ s

wan2.6-i2v

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.5-i2v-preview

International

Video with audio

480P

$0.05/ s

None

720P

$0.10/ s

1080P

$0.15/ s

wan2.2-i2v-flash

International

Silent video

480P

$0.015/ s

None

720P

$0.036/ s

wan2.2-i2v-plus

International

Silent video

480P

$0.02/ s

None

1080P

$0.10/ s

wan2.1-t2v-turbo

International

Silent video

480P

$0.036/ s

50 s

720P

$0.036/ s

wan2.1-t2v-plus

International

Silent video

720P

$0.10/ s

50 s

China (Beijing)

Model ID

Output video type

Output video resolution

Output unit price

wan2.6-i2v-flash

Video with audio

audio=true

720P

$0.043006/ s

1080P

$0.071676/ s

Silent video

audio=false

720P

$0.021503/ s

1080P

$0.035838/ s

wan2.6-i2v

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.5-i2v-preview

Video with audio

480P

$0.043006/ s

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.2-i2v-plus

Silent video

480P

$0.02007/ s

1080P

$0.100347/ s

wanx2.1-t2v-turbo

Silent video

480P

$0.034405/ s

720P

$0.034405/ s

wanx2.1-t2v-plus

Silent video

720P

$0.100347/ s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video type

Output video resolution

Output unit price

wan2.6-i2v

Global

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

United States (Virginia)

Model ID

Service Deployment Scope

Output video type

Output video resolution

Output unit price

wan2.6-i2v

Global

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.6-i2v

United States

Video with audio

720P

$0.1/ s

1080P

$0.15/ s

Wan-I2V based on first and last frames

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution; failed requests incur no fees and do not consume the free quota.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.2-kf2v-flash

International

480P

$0.015/ s

None

720P

$0.036/ s

1080P

$0.07/s

wan2.1-kf2v-plus

International

720P

$0.10/ s

50 s

China (Beijing)

Model ID

Output video resolution

Output unit price

wan2.2-kf2v-flash

480P

$0.014335/ s

720P

$0.028671/ s

1080P

$0.068809/ s

wanx2.1-kf2v-plus

720P

$0.100347/ s

Wan-reference-to-video

Billing rule: Both input videos and output videos are billed by video seconds.

Failed requests do not incur fees or consume the free quota.

Billing formula: Billable duration = Input video duration (up to 5 seconds) + Output video duration.

  • The billable duration of the input video does not exceed 5 seconds. For the calculation rules, see Billing and rate limits.
  • The billable duration of the output video is the number of seconds of successfully generated video.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Output video type

Output video resolution

Input and output unit prices

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.7-r2v-2026-06-12

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.7-r2v

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

wan2.6-r2v-flash

International

Video with audio

audio=true

720P

$0.05/ s

None

1080P

$0.075/ s

Silent video

audio=false

720P

$0.025/ s

1080P

$0.0375/ s

wan2.6-r2v

International

Video with audio

720P

$0.10/ s

None

1080P

$0.15/ s

China (Beijing)

Model ID

Output video type

Output video resolution

Input and output unit prices

wan2.7-r2v-2026-06-12

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.7-r2v

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

wan2.6-r2v-flash

Video with audio

audio=true

720P

$0.043006/ s

1080P

$0.071676/ s

Silent video

audio=false

720P

$0.021503/ s

1080P

$0.035838/ s

wan2.6-r2v

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

Germany (Frankfurt)

Model ID

Service Deployment Scope

Output video type

Output video resolution

Input and output unit prices

wan2.6-r2v

Global

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

United States (Virginia)

Model ID

Service Deployment Scope

Output video type

Output video resolution

Input and output unit prices

wan2.6-r2v

Global

Video with audio

720P

$0.086012/ s

1080P

$0.143353/ s

Wan-Video Editing

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Billing rule: For wan2.7-videoedit, both the input video and the output video are billed by video seconds; for wan2.1-vace-plus (wanx2.1-vace-plus in the China (Beijing) region), only the output video is billed by video seconds.

Failed requests do not incur fees or consume the free quota.

Singapore

Model ID

Service Deployment Scope

Output video resolution

Input and output unit prices

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.7-videoedit

International

720P

$0.10/ s

None

1080P

$0.15/ s

Model ID

Service Deployment Scope

Output video resolution

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.1-vace-plus

International

720P

$0.10/ s

50 s

China (Beijing)

Model ID

Output video resolution

Input and output unit prices

wan2.7-videoedit

720P

$0.086012/ s

1080P

$0.143353/ s

Model ID

Service Deployment Scope

Output video resolution

Output unit price

wanx2.1-vace-plus

the Chinese mainland

720P

$0.100347/ s

Wan-Digital Human

  • wan2.2-s2v-detect: The input is billed and the output is not billed. The input is billed based on the number of images detected; as long as the request succeeds (regardless of whether the detection result passes), each input image is billed once.
  • wan2.2-s2v: The input is not billed, and the output is billed based on the number of seconds of successfully generated video. Unit prices vary by output video resolution (480P/720P); failed requests incur no fees and do not consume the free quota.

China (Beijing)

Model ID

Unit price

Free quota(Note)

wan2.2-s2v-detect

Input image: $0.000574/image

None

wan2.2-s2v

Output video:

  • 480P: $0.071677/ s

  • 720P: $0.129018/ s

None

Wanx-Image-to-Action

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

The unit price varies by output video mode (standard mode/professional mode); failed requests do not incur fees or consume the free quota.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Output video mode

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.2-animate-move

International

Standard modewan-std

$0.12/ s

None

Professional modewan-pro

$0.18/s

China (Beijing)

Model ID

Output video mode

Output unit price

wan2.2-animate-move

Standard modewan-std

$0.06/ s

Professional modewan-pro

$0.09/ s

Wanx-Video Face Swap

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

The unit price varies by output video mode (standard mode/professional mode); failed requests do not incur fees or consume the free quota.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Output video mode

Output unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

wan2.2-animate-mix

International

Standard modewan-std

$0.18/s

None

Professional modewan-pro

$0.26/ s

China (Beijing)

Model ID

Output video mode

Output unit price

wan2.2-animate-mix

Standard modewan-std

$0.09/ s

Professional modewan-pro

$0.13/ s

Talking Portrait VideoRetalk

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Uniform pricing applies; failed requests do not incur fees or consume the free quota.

China (Beijing)

Model ID

Output unit price

Free quota(Note)

videoretalk

$0.011469/ s

None

Video Style Repainting

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

The unit price varies by output video resolution (540P/720P); failed requests do not incur fees or consume the free quota.

China (Beijing)

Model ID

Output video resolution

Output unit price

Free quota(Note)

video-style-transform

540P

$0.028671/ s

None

720P

$0.071677/ s

Vidu-Reference-to-Video

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution; failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video resolution

Output unit price

Free quota(Note)

vidu/viduq3-mix_reference2video

720P

$0.12/ s

None

1080P

$0.145/ s

vidu/viduq3-ad_reference2video

720P

$0.125/ s

None

1080P

$0.15/ s

vidu/viduq3-drama_reference2video

720P

$0.14/s

None

1080P

$0.14/s

Vidu-I2V-First Frame

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution; failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video resolution

Output unit price

Free quota(Note)

vidu/viduq2-pro-fast_img2video

720P

$0.016/ s

None

1080P

$0.032/ s

Aishi-T2V

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution and video type (with audio/silent); failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video type

Output video resolution

Output unit price

Free quota(Note)

pixverse/pixverse-v6-t2v

Video with audio

audio=true

360P

$0.032/ s

None

540P

$0.04/ s

720P

$0.054/ s

1080P

$0.103/ s

Silent video

audio=false

360P

$0.023/ s

540P

$0.032/ s

720P

$0.04/ s

1080P

$0.08/ s

Aishi-I2V-First-Frame

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution and video type (with audio/silent); failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video type

Output video resolution

Output unit price

Free quota(Note)

pixverse/pixverse-v6-it2v

Video with audio

audio=true

360P

$0.032/ s

None

540P

$0.04/ s

720P

$0.054/ s

1080P

$0.103/ s

Silent video

audio=false

360P

$0.023/ s

540P

$0.032/ s

720P

$0.04/ s

1080P

$0.08/ s

Aishi-I2V-First-and-Last-Frame

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution and video type (with audio/silent); failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video type

Output video resolution

Output unit price

Free quota(Note)

pixverse/pixverse-v6-kf2v

Video with audio

audio=true

360P

$0.032/ s

None

540P

$0.04/ s

720P

$0.054/ s

1080P

$0.103/ s

Silent video

audio=false

360P

$0.023/ s

540P

$0.032/ s

720P

$0.04/ s

1080P

$0.08/ s

Aishi-Reference-to-Video

Billing rule: Input is not billed; output is billed by the seconds of video successfully generated.

Unit prices vary by output video resolution and video type (with audio/silent); failed requests incur no fees and do not consume the free quota.

Singapore

Model ID

Output video type

Output video resolution

Output unit price

Free quota(Note)

pixverse/pixverse-v6-r2v

Video with audio

audio=true

360P

$0.032/ s

None

540P

$0.04/ s

720P

$0.054/ s

1080P

$0.103/ s

Silent video

audio=false

360P

$0.023/ s

540P

$0.032/ s

720P

$0.04/ s

1080P

$0.08/ s

World Model

HappyOyster-World Exploration (Adventure)

Billing rule:

  • World creation: Billed by the number of world creations, Fee = Number of world creations × Unit price.
  • World experience: Billed by the duration (seconds) of output videos, Fee = Video duration × Unit price, only 480P is supported.
  • Failed requests do not incur any fees.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-adventure

International

World creation

$0.0077/Times

World experience (480P)

$0.0308/ s

China (Beijing)

Model ID

Billing item

Unit price

happyoyster-1.0-adventure

World creation

$0.007067/Times

World experience (480P)

$0.028267/ s

United States (Virginia)

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-adventure

Global

World creation

$0.007067/Times

World experience (480P)

$0.028267/ s

HappyOyster-Real-Time Directing (Directing)

Billing rule:

  • World creation: Billed by the number of world creations, Fee = Number of world creations × Unit price. Whether reference images are used affects the unit price: with and without reference images are billed separately.
  • Instruction injection: Billed by the number of instructions injected, Fee = Number of instructions × Unit price.
  • World experience: Billed by the duration (seconds) of output videos, Fee = Video duration × Unit price, unit prices vary by resolution (480P/720P).
  • Failed requests do not incur any fees.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-directing

International

World creation (without reference image)

$0.3113/Times

World creation (with reference image)

$0.0077/Times

Instruction injection

$0.0154/Times

World experience (480P)

$0.0539/ s

World experience (720P)

$0.0858/ s

United States (Virginia)

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-directing

Global

World creation (without reference image)

$0.282666/Times

World creation (with reference image)

$0.007067/Times

Instruction injection

$0.014133/Times

World experience (480P)

$0.049466/ s

World experience (720P)

$0.077733/ s

HappyOyster-Role Acting

Billing rule:

  • World creation: Billed based on the number of world creations, Fee = Number of world creations × unit price.
  • Instruction injection: Billed based on the number of injected instructions, Fee = Number of instructions × unit price.
  • World experience: Billed based on the duration of output videos (in seconds), Fee = Video Duration × unit price; unit prices vary by resolution (480P/720P).
  • Failed requests do not incur any fees.

NoteThe following models have no free quota in any region.

Singapore

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-acting
(invite-only beta)

International

World creation

$0.0077/Times

Instruction injection

$0.0077/Times

World experience (480P)

$0.0308/ s

World experience (720P)

$0.0462/s

United States (Virginia)

Model ID

Service Deployment Scope

Billing item

Unit price

happyoyster-1.0-acting
(invite-only beta)

Global

World creation

$0.007067/Times

Instruction injection

$0.007067/Times

World experience (480P)

$0.028267/ s

World experience (720P)

$0.0424/s

Speech recognition

Qwen-Audio-ASR

Billing rule:

  • qwen-audio-3.1-asr-flash: Billed separately based on the number of input and output tokens.
  • qwen-audio-3.0-asr-flash: Billed based on the duration (in seconds) of input audio; output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit priceOutput unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-audio-3.1-asr-flashInternational0.15 USD/million tokens0.47 USD/million tokens1 million tokens
qwen-audio-3.0-asr-flashInternational$0.000035/sNot billed

None

China (Beijing)

Model IDInput unit priceOutput unit price
qwen-audio-3.1-asr-flash0.113 USD/million tokens0.382 USD/million tokens
qwen-audio-3.0-asr-flash$0.00003/sNot billed

Qwen-Audio-ASR-Streaming

Billing rule:

  • qwen-audio-3.1-asr-flash-streaming: Billed separately based on the number of input and output tokens.
  • qwen-audio-3.0-asr-flash-streaming: Billed based on the duration (in seconds) of input audio; output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit priceOutput unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-audio-3.1-asr-flash-streamingInternational0.93 USD/million tokens0.70 USD/million tokens1 million tokens
qwen-audio-3.0-asr-flash-streamingInternational$0.00009/sNot billed

None

China (Beijing)

Model IDInput unit priceOutput unit price
qwen-audio-3.1-asr-flash-streaming0.848 USD/million tokens0.636 USD/million tokens
qwen-audio-3.0-asr-flash-streaming$0.000047/sNot billed

Qwen-Audio-ASR-Filetrans

Billing rule:

  • qwen-audio-3.1-asr-flash-filetrans: Billed separately based on the number of input and output tokens.
  • qwen-audio-3.0-asr-flash-filetrans: Billed based on the duration (in seconds) of input audio; output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit priceOutput unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-audio-3.1-asr-flash-filetransInternational0.15 USD/million tokens0.47 USD/million tokens1 million tokens
qwen-audio-3.0-asr-flash-filetransInternational$0.000035/sNot billed

None

China (Beijing)

Model IDInput unit priceOutput unit price
qwen-audio-3.1-asr-flash-filetrans0.113 USD/million tokens0.382 USD/million tokens
qwen-audio-3.0-asr-flash-filetrans$0.000032/sNot billed

Qwen-Audio-ASR-Message

Billing rule: Billed based on the number of input and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-audio-3.1-asr-flash-messageInternational0.93</td><tdstyle=verticalAlign:"top">0.93</td><td style={{ verticalAlign: "top" }}>0.701 million Tokens

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)
qwen-audio-3.1-asr-flash-message0.848</td><tdstyle=verticalAlign:"top">0.848</td><td style={{ verticalAlign: "top" }}>0.636

Qwen ASR

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-asr-flash-filetrans

International

$0.000035/s

36,000 seconds (10 hours)

qwen3-asr-flash

International

$0.000035/s

36,000 seconds (10 hours)

China (Beijing)

Model IDInput unit price

qwen3-asr-flash-filetrans

$0.000032/s

qwen3-asr-flash

$0.000032/s

United States (Virginia)

Model ID

Service Deployment Scope

Input unit price

qwen3-asr-flash

United States

$0.000035/s

Qwen ASR-Realtime

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-asr-flash-realtime

International

$0.000090/s

36,000 seconds (10 hours)

China (Beijing)

Model IDInput unit price

qwen3-asr-flash-realtime

$0.000047/s

Fun-ASR

File transcription

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

fun-asr

International

$0.000035/s

36,000 seconds (10 hours)

fun-asr-flash-2026-06-15

International

$0.000035/s

36,000 seconds (10 hours)

China (Beijing)

Model IDInput unit price

fun-asr

$0.000032/s

fun-asr-flash-2026-06-15

$0.00003/s

Real-time Speech Recognition

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

Singapore

Model ID

Service Deployment Scope

Input unit price

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

fun-asr-realtime

International

$0.00009/s

36,000 seconds (10 hours)

China (Beijing)

Model IDInput unit price

fun-asr-realtime

$0.000047/s

fun-asr-realtime-2026-02-28

$0.000047/s

fun-asr-mtl-realtime-2025-12-10

$0.000047/s

fun-asr-flash-8k-realtime

The current capability is equivalent to fun-asr-flash-8k-realtime-2026-01-28

$0.000032/s

fun-asr-flash-8k-realtime-2026-01-28

$0.000032/s

Paraformer

File transcription

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

China (Beijing)

Model ID

Input unit price

paraformer-v2

$0.000012/s

paraformer-8k-v2

$0.000012/s

Real-time Speech Recognition

Billing rule: You are billed based on the number of seconds of the input audio. Output is not billed.

China (Beijing)

Model ID

Input unit price

Free quota(Note)

paraformer-realtime-v2

$0.000035/s

None

paraformer-realtime-8k-v2

$0.000035/s

None

Speech Synthesis

Character counting rules: For models in this section that are billed by character count (input unit price is per "10,000 characters"), the number of characters in the input text is counted according to the following rules:

  • A Chinese character (including Simplified Chinese characters, Traditional Chinese characters, Japanese Kanji, and Korean Hanja) counts as 2 characters.
  • Other characters (such as an English letter, a digit, a punctuation mark, a space, a Japanese kana, and a Korean letter) count as 1 character.
  • When SSML is used, the SSML tags themselves are not counted; only the text content to be synthesized is counted.

Examples: Two Chinese characters count as 4 characters (2+2); two Chinese characters plus "A123" count as 8 characters (2+1+2+1+1+1); two Chinese characters plus a period count as 5 characters (2+2+1); two Chinese characters, a space, and a period count as 6 characters (2+1+2+1).

Qwen-Audio-TTS

Billing rule: qwen-audio-3.1-tts-flash is billed separately based on the number of input tokens and output tokens; other models are billed based on the number of characters in the input text, and output is not billed.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-audio-3.1-tts-flash0.23</td><tdstyle=verticalAlign:"top">0.23</td><td style={{ verticalAlign: "top" }}>1.871 million tokens
Model IDService Deployment ScopeInput unit price (per 10,000 characters)Free quota(Note)Validity: Within 90 days from the date of activating Model Studio/model release/application approval (whichever is later)

qwen-audio-3.1-tts-plus

International

$0.3

10,000 characters

qwen-audio-3.0-tts-plus

International

$0.2

None

qwen-audio-3.0-tts-flash

International

$0.15

None

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)
qwen-audio-3.1-tts-flash0.212 USD1.696 USD
Model IDInput unit price (per 10,000 characters)

Free quota (Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen-audio-3.1-tts-plus

$0.2

qwen-audio-3.0-tts-plus

$0.19253

qwen-audio-3.0-tts-flash

$0.137521

CosyVoice

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model ID

Service Deployment Scope

Input unit price (per 10,000 characters)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

cosyvoice-v3-plus

International

$0.26

10,000 characters

cosyvoice-v3-flash

International

$0.13

10,000 characters

China (Beijing)

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model ID

Input unit price (per 10,000 characters)

cosyvoice-v3.5-plus

$0.22

cosyvoice-v3.5-flash

$0.116

cosyvoice-v3-plus

$0.286706

cosyvoice-v3-flash

$0.14335

cosyvoice-v2

$0.286706

Qwen-TTS

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Qwen3-TTS-Flash

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model IDService Deployment ScopeInput unit price (per 10,000 characters)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-tts-flash

International

$0.1

10,000 characters

China (Beijing)

Qwen3-TTS-Flash

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model IDService Deployment ScopeInput unit price (per 10,000 characters)Output unit price (per 10,000 characters)

qwen3-tts-flash

the Chinese mainland

$0.114682

Not billed

Qwen-TTS

Billing rule: Billing is based on input tokens and output tokens.

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

qwen-tts-flash

the Chinese mainland

$0.23

$1.434

Qwen-TTS-Realtime

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Qwen3-TTS-Flash-Realtime

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model IDService Deployment ScopeInput unit price (per 10,000 characters)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen3-tts-flash-realtime

International

$0.13

Alibaba Cloud Model Studio activated after 00:00 on November 13, 2025: 10,000 characters

China (Beijing)

Qwen3-TTS-Flash-Realtime

Billing rule: Fees are based on the number of characters in the input text. Output is not billed.

Model ID

Service Deployment Scope

Input unit price (per 10,000 characters)

Output unit price

qwen3-tts-flash-realtime

the Chinese mainland

$0.143353

Not billed

Audio Generation

Qwen-Audio-TTS-Next

Input and output tokens are billed separately.

Singapore

Model IDInput unit price (USD/million tokens)Output unit price (USD/million tokens)Free quota (Note)
Validity: Within 90 days from the date of activating Model Studio/model release/application approval (whichever is later)
qwen-audio-3.1-tts-next0.931.871 million tokens

China (Beijing)

Model IDInput unit price (USD/million tokens)Output unit price (USD/million tokens)
qwen-audio-3.1-tts-next0.8481.696

Voice Chat

Qwen-Audio-Realtime

The realtime voice chat model supports text and audio input and output, and bills input tokens and output tokens separately. Audio tokens are converted based on duration: Total number of tokens = Audio duration (in seconds) * 12.5. Audio under 1 second is counted as 1 second.

When only text is output, fees are charged at the text output unit price; when both text and audio are output, only the audio incurs output fees, and the accompanying text is not charged additional output fees.

In multi-turn conversation scenarios, consistent with text LLMs, the model maintains the complete conversation context to preserve coherent conversation capability. Historical conversation content is processed and billed as input for subsequent turns, so the number of input tokens per turn grows gradually as the number of conversation turns increases. The specific billing methods for each type of content are as follows:

  • User input audio and text: Counted in the context and billed as input in each subsequent turn. Audio is billed as audio tokens, and text is billed as text tokens.
  • User-set instructions: Billed once per turn as text tokens.
  • Model output text: Counted in the context as text tokens and billed as input in each subsequent turn.
  • Model output audio: Billed as audio tokens only once at output, and not counted in the context.

As the number of conversation turns increases, the number of tokens accumulated in the context grows gradually. We recommend that you reasonably control the number of turns per session or start a new session at an appropriate time to optimize usage costs.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

Text

Audio

Text

Audio

qwen-audio-3.1-realtime-plus

International

$0.8

$6.4

$6.4

$24

1 million Tokens

qwen-audio-3.1-realtime-flash

International

$0.23

$0.93

$0.7

$1.87

1 million Tokens

qwen-audio-3.0-realtime-plus

International

$0.8

$6.4

$6.4

$24

None

qwen-audio-3.0-realtime-flash

International

$0.23

$0.93

$0.7

$1.87

None

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Text

Audio

Text

Audio

qwen-audio-3.1-realtime-plus

$0.707

$5.653

$5.653

$21.2

qwen-audio-3.1-realtime-flash

$0.212

$0.848

$0.636

$1.696

qwen-audio-3.0-realtime-plus

$0.688

$5.501

$5.501

$20.628

qwen-audio-3.0-realtime-flash

$0.212

$0.848

$0.636

$1.696

Audio and Video Translation

Qwen-LiveTranslate-Flash-Realtime

Billing rule: Fees are based on input tokens and output tokens. For token calculation rules for different modalities, see Billing Details.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)
Input: AudioInput: ImageOutput: TextOutput: Audio

qwen3.8-livetranslate-flash-realtime

International

$7.5

$0.55

$20

$30

1 million Tokens

qwen3.5-livetranslate-flash-realtime

International

$7.5

$0.55

$20

$30

None

qwen3.5-livetranslate-flash-realtime-2026-05-19

International

$7.5

$0.55

$20

$30

None

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)
Input: AudioInput: ImageOutput: TextOutput: Audio

qwen3.8-livetranslate-flash-realtime

$5.653

$0.466

$14.133

$22.613

qwen3.5-livetranslate-flash-realtime

$5.501

$0.454

$13.752

$22.003

qwen3.5-livetranslate-flash-realtime-2026-05-19

$5.501

$0.454

$13.752

$22.003

Qwen-LiveTranslate-Flash

Billing rule: Fees are based on input tokens and output tokens. For token calculation rules for different modalities, see Billing Details.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

Input: Audio

Input: Image

Output: Text

Output: Audio

qwen3-livetranslate-flash

International

$1.577

$0.631

$1.577

$6.308

1 million Tokens

qwen3-livetranslate-flash-2025-12-01

International

$1.577

$0.631

$1.577

$6.308

None

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Input: Audio

Input: Image

Output: Text

Output: Audio

qwen3-livetranslate-flash

$1.434

$0.573

$1.434

$5.734

qwen3-livetranslate-flash-2025-12-01

$1.434

$0.573

$1.434

$5.734

Voice Cloning/Design

Qwen-TTS Voice Cloning

Billing rule: Fees are charged based on the number of newly created voices.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model ID

Service Deployment Scope

Unit price (per voice)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen-voice-enrollment

International

$0.01

1,000 voices/account

China (Beijing)

Model ID

Unit price (per voice)

qwen-voice-enrollment

$0.01

Music Generation

Fun-Music

Billing rule: Fees are charged based on the number of seconds of output audio. Input is not charged.

China (Beijing)

Model ID

Output unit price (per second)

Free quota(Note)

fun-music-preview

$0.000695

None

fun-music-v1

$0.000275

None

Text Embedding

Billing rule: Fees are charged based on input tokens. Output is not charged.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen3.7-text-embedding

International

$0.07

1 million Tokens

text-embedding-v4

International

$0.07

1 million Tokens

text-embedding-v3

International

$0.07

None

China (Beijing)

Model ID

Input unit price (per million tokens)

text-embedding-v4

$0.072

Hong Kong (China)

Model ID

Service Deployment Scope

Input unit price (per million tokens)

text-embedding-v4

Hong Kong (China)

$0.07

Multimodal Embedding

Billing rule: Fees are charged based on input tokens. Output is not charged.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million input tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

tongyi-embedding-vision-plus

International

$0.09

1 million Tokens

tongyi-embedding-vision-flash

International

Image/Video: $0.03

Text: $0.09

1 million Tokens

China (Beijing)

Model ID

Input unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen3-vl-embedding

Image/Video: $0.258

Text: $0.1

1 million Tokens

multimodal-embedding-v1

Free trial

No token quota limit

Reranking Model

Billing rule: Fees are charged based on input tokens. Output is not charged.

Singapore

Model ID

Service Deployment Scope

Input unit price (per million tokens)

Free quota(Note)

Validity: 90 days from the date you activate Model Studio, the model is released, or your application is approved (whichever is later)

qwen3-rerank

International

$0.1

1 million Tokens

China (Beijing)

Model ID

Input unit price (per million tokens)

qwen3-rerank

Text input: $0.069

qwen3-vl-rerank

Text input: $0.1

Image input: $0.258

gte-rerank-v2

Text input: $0.115

Industry Models

Intention Recognition

China (Beijing)

Model ID

Input unit price (per million tokens)

Output unit price (per million tokens)

Free quota(Note)

tongyi-intent-detect-v3

$0.058

$0.144

None

Role Assumption

Billing rule: Billing is based on input tokens and output tokens.

NoteThe following models offer a free quota only in the Singapore region, and no free quota is available in other regions.

Singapore

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

Free quota(Note)

Validity: Within 90 days from the date of Model Studio activation/model release/application approval (whichever is later)

qwen-plus-character

Session Cache is eligible for a discount

International

$0.5

$1.4

1 million Tokens

qwen-flash-character

Session Cache is eligible for a discount

International

$0.05

$0.4

1 million Tokens

qwen-plus-character-ja

International

$0.5

$1.4

1 million Tokens

China (Beijing)

Model IDInput unit price (per million tokens)Output unit price (per million tokens)

qwen-plus-character

Session Cache is eligible for a discount

$0.115

$0.287

qwen-flash-character

Session Cache is eligible for a discount

$0.034

$0.203

Hong Kong (China)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-plus-character

Session Cache is eligible for a discount

Global

$0.115

$0.287

Germany (Frankfurt)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-plus-character

Session Cache is eligible for a discount

Global

$0.115

$0.287

United States (Virginia)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-plus-character

Session Cache is eligible for a discount

Global

$0.115

$0.287

qwen-flash-character

Session Cache is eligible for a discount

Global

$0.034

$0.203

Japan (Tokyo)

Model IDService Deployment ScopeInput unit price (per million tokens)Output unit price (per million tokens)

qwen-plus-character

Session Cache is eligible for a discount

Global

$0.115

$0.287

Decision Model

Billing rule: Fees are charged based on input tokens.

Singapore

Model ID

Input unit price (per million tokens)

Free quota

decision-model-preview

Limited Time Free

China (Beijing)

Model ID

Input unit price (per million tokens)

Free quota

decision-model-preview

Limited Time Free

Token Consumption and Cost Control

Billing method for reading files via URL: When reading files via URL, the file transfer process itself does not consume tokens. After the file content is parsed, it is converted into input tokens for billing, and the consumption depends on the length of the parsed text, not the file size. If you need to process large files, we recommend using knowledge base chunked indexing to reduce token consumption.

How to troubleshoot abnormally high Credits consumption in a single request: Common causes include an excessively large input token count, enabling Agent mode (which incurs substantial overhead from system prompts, tool definitions, and thinking content), and so on. You can optimize costs by compressing the conversation history, starting a new conversation, disabling thinking mode, or switching to a lightweight model (such as the Flash series).

Recommended models for low-frequency call scenarios: For low-frequency call scenarios such as heartbeat checks, we recommend using low-cost models such as the Qwen Flash series to reduce overall call costs.

Note on decommissioned models: Decommissioned models such as the Qwen2.5 series are no longer available, and their prices cannot be queried. Please use newer models instead.

Error code

If the model call fails and returns an error message, see Error code for troubleshooting.