qwen3.8-max

Updated at:

A 2.4-trillion-parameter MoE flagship model with improved capabilities in coding and office scenarios, suitable for professional tasks such as law, finance, and design. It provides native visual understanding, supports semantic parsing of ultra-long documents and long videos, and supports autonomous planning and iteration for long-horizon tasks.

Model ID (the value of the model parameter): qwen3.8-max

Snapshot: qwen3.8-max-0902 (alias qwen3.8-max-2026-09-02)

Model Capabilities

China (Beijing)

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Supported

Fine-tuning

Unsupported

Singapore

Scope: International

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Germany (Frankfurt)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Unsupported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

US (Virginia)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Unsupported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Japan (Tokyo)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Unsupported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Hong Kong (China)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Unsupported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter

Value

Parameter

Value

Max Input Length

991808

Max Output Length

131072

Max Input Length (Thinking Mode)

983616

Max Output Length (Thinking Mode)

131072

Context Window

1000000

Max Chain-of-Thought Length

262144

The supported model length may vary depending on different combinations of API input parameters.

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Input(Batch File)

6

Per 1M tokens

Output(Batch File)

18

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Input(Batch Chat)

12

Per 1M tokens

Output(Batch Chat)

36

Per 1M tokens

Singapore

Scope: International

Billing Item

Price (CNY)

Unit

Input

14.988

Per 1M tokens

Output

44.965

Per 1M tokens

Input(Implicit Cache)

1.874

Per 1M tokens

Explicit Cache Creation

18.736

Per 1M tokens

Explicit Cache Read

1.274

Per 1M tokens

Germany (Frankfurt)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

US (Virginia)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Japan (Tokyo)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Rate Limits

NoteThis model uses dynamic rate limiting in the China (Beijing) and Singapore regions. TPM is adjusted based on your monthly Model Studio spend tier, while RPM has a relatively high limit that you do not need to configure separately and that will not be triggered during normal use.

China (Beijing)

This region uses dynamic rate limiting. The TPM limit is tiered by your monthly Model Studio spend, and the RPM limit is high enough that normal use does not trigger throttling. For the tier values, see Dynamic rate limiting.

Singapore

Scope: International

This region uses dynamic rate limiting. The TPM limit is tiered by your monthly Model Studio spend, and the RPM limit is high enough that normal use does not trigger throttling. For the tier values, see Dynamic rate limiting.

Germany (Frankfurt)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

5,000,000

US (Virginia)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

5,000,000

Japan (Tokyo)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

5,000,000

Hong Kong (China)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

5,000,000

Snapshot Versions

qwen3.8-max-0902

Qwen3.8-Max-0902 (alias qwen3.8-max-2026-09-02) is an upgraded snapshot of qwen3.8-max. Coding capability breaks new ground, handling more complex engineering-scale projects and long-horizon autonomous development. Collaborative agent performance is significantly enhanced, with greater composure in multi-tool orchestration and end-to-end task delivery. Native vision understanding is refined across chart reasoning, document parsing, and multimodal perception — sharper and more reliable. Retains the 1M context window, thinking mode, and full tool ecosystem, evolving at a higher level of intelligence.

Model Capabilities

China (Beijing)

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Singapore

Scope: International

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Germany (Frankfurt)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

US (Virginia)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Japan (Tokyo)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Hong Kong (China)

Scope: Global

Capability

Support

Capability

Support

Input Modality

Image Text Video

Output Modality

Text

Model Experience

Supported

Function Calling

Supported

Structured Outputs

Supported

Web Search

Supported

Prefix Completion

Supported

Context Caching

Supported

Batch Inference

Unsupported

Fine-tuning

Unsupported

Context Limits

Parameter

Value

Parameter

Value

Max Input Length

991808

Max Output Length

131072

Max Input Length (Thinking Mode)

983616

Max Output Length (Thinking Mode)

131072

Context Window

1000000

Max Chain-of-Thought Length

262144

The supported model length may vary depending on different combinations of API input parameters.

Pricing

This page only shows the original pricing for model API calls, excluding any limited-time promotions. Visit Model Studio Console for promotional offers.

China (Beijing)

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Singapore

Scope: International

Billing Item

Price (CNY)

Unit

Input

14.988

Per 1M tokens

Output

44.965

Per 1M tokens

Input(Implicit Cache)

1.874

Per 1M tokens

Explicit Cache Creation

18.736

Per 1M tokens

Explicit Cache Read

1.274

Per 1M tokens

Germany (Frankfurt)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

US (Virginia)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Japan (Tokyo)

Scope: Global

Billing Item

Price (CNY)

Unit

Input

12

Per 1M tokens

Output

36

Per 1M tokens

Input(Implicit Cache)

1.5

Per 1M tokens

Explicit Cache Creation

15

Per 1M tokens

Explicit Cache Read

1

Per 1M tokens

Rate Limits

NoteThis model uses dynamic rate limiting in the China (Beijing) and Singapore regions. TPM is adjusted based on your monthly Model Studio spend tier, while RPM has a relatively high limit that you do not need to configure separately and that will not be triggered during normal use.

China (Beijing)

This region uses dynamic rate limiting. The TPM limit is tiered by your monthly Model Studio spend, and the RPM limit is high enough that normal use does not trigger throttling. For the tier values, see Dynamic rate limiting.

Singapore

Scope: International

This region uses dynamic rate limiting. The TPM limit is tiered by your monthly Model Studio spend, and the RPM limit is high enough that normal use does not trigger throttling. For the tier values, see Dynamic rate limiting.

Germany (Frankfurt)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

150,000

US (Virginia)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

150,000

Japan (Tokyo)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

150,000

Hong Kong (China)

Scope: Global

Parameter

Value

RPM (Requests Per Minute)

30,000

TPM (Tokens Per Minute)

150,000