The Qwen 3 series Max model has undergone specialized upgrades in agent programming and tool invocation compared to the preview version. The officially released model this time has achieved state-of-the-art (SOTA) performance in its field and is better suited to meet the demands of agents operating in more complex scenarios.This model version is functionally equivalent to the snapshot model qwen3-max-2026-01-23.
Inference Service Provider
The inference service provider for qwen3-max is Alibaba Cloud Model Studio.
Model Capabilities
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
- Hong Kong (China)
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | 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 |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 258048 | Max Output Length | 65536 |
Context Window | 262144 | Max Input Length (Thinking Mode) | 258048 |
Max Output Length (Thinking Mode) | 32768 | Max Chain-of-Thought Length | 81920 |
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)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
- Hong Kong (China)
Input<=32k
Input<=32k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.359 | Per 1M tokens |
Output | 1.434 | Per 1M tokens |
Input(Implicit Cache) | 0.072 | Per 1M tokens |
Input(Batch File) | 0.18 | Per 1M tokens |
Output(Batch File) | 0.717 | Per 1M tokens |
Explicit Cache Creation | 0.448 | Per 1M tokens |
Explicit Cache Read | 0.036 | Per 1M tokens |
Input(Batch Chat) | 0.359 | Per 1M tokens |
Output(Batch Chat) | 1.434 | Per 1M tokens |
32k<Input<=128k
32k<Input<=128k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.574 | Per 1M tokens |
Output | 2.294 | Per 1M tokens |
Input(Implicit Cache) | 0.115 | Per 1M tokens |
Input(Batch File) | 0.287 | Per 1M tokens |
Output(Batch File) | 1.147 | Per 1M tokens |
Explicit Cache Creation | 0.717 | Per 1M tokens |
Explicit Cache Read | 0.058 | Per 1M tokens |
Input(Batch Chat) | 0.574 | Per 1M tokens |
Output(Batch Chat) | 2.294 | Per 1M tokens |
128k<Input<=256k
128k<Input<=256k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 1.004 | Per 1M tokens |
Output | 4.014 | Per 1M tokens |
Input(Implicit Cache) | 0.201 | Per 1M tokens |
Input(Batch File) | 0.502 | Per 1M tokens |
Output(Batch File) | 2.007 | Per 1M tokens |
Explicit Cache Creation | 1.255 | Per 1M tokens |
Explicit Cache Read | 0.101 | Per 1M tokens |
Input(Batch Chat) | 1.004 | Per 1M tokens |
Output(Batch Chat) | 4.014 | Per 1M tokens |
Rate Limits
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
- Hong Kong (China)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 30000 |
TPM (Tokens Per Minute) | 5,000,000 |
Snapshot Versions
qwen3-max-preview
A preview version of the Max model in the Qwen 3 series, achieving an effective integration of thinking and non-thinking modes. In thinking mode, there is a significant enhancement in capabilities such as intelligent agent programming, common-sense reasoning, and reasoning across mathematics, science, and general domains.
Inference Service Provider
The inference service provider for qwen3-max-preview is Alibaba Cloud Model Studio.
Model Capabilities
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | Output Modality | Text |
Model Experience | Supported | Function Calling | Unsupported |
Structured Outputs | Unsupported | Web Search | Unsupported |
Prefix Completion | Unsupported | Context Caching | Supported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 258048 | Max Output Length | 65536 |
Context Window | 262144 | Max Input Length (Thinking Mode) | 258048 |
Max Output Length (Thinking Mode) | 32768 | Max Chain-of-Thought Length | 81920 |
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)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
Input<=32k
Input<=32k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.861 | Per 1M tokens |
Output | 3.441 | Per 1M tokens |
Input(Implicit Cache) | 0.173 | Per 1M tokens |
32k<Input<=128k
32k<Input<=128k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 1.434 | Per 1M tokens |
Output | 5.735 | Per 1M tokens |
Input(Implicit Cache) | 0.287 | Per 1M tokens |
128k<Input<=256k
128k<Input<=256k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 2.151 | Per 1M tokens |
Output | 8.602 | Per 1M tokens |
Input(Implicit Cache) | 0.431 | Per 1M tokens |
Rate Limits
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 600 |
TPM (Tokens Per Minute) | 1,000,000 |
qwen3-max-2026-01-23
Compared with the snapshot as of September 23, 2025, the Qwen-3 series Max model in this release achieves an effective integration of thinking and non-thinking modes, resulting in a comprehensive and substantial improvement in the model’s overall performance. In thinking mode, the model simultaneously supports web search, web information extraction, and a code interpreter tool, enabling it to tackle more complex and challenging problems with greater accuracy by leveraging external tools while engaging in slow, deliberative reasoning. This version is based on a snapshot taken on January 23, 2026.
Inference Service Provider
The inference service provider for qwen3-max-2026-01-23 is Alibaba Cloud Model Studio.
Model Capabilities
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- Hong Kong (China)
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | Output Modality | Text |
Model Experience | Supported | Function Calling | Supported |
Structured Outputs | Supported | Web Search | Supported |
Prefix Completion | Supported | Context Caching | Unsupported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 258048 | Max Output Length | 65536 |
Context Window | 262144 | Max Input Length (Thinking Mode) | 258048 |
Max Output Length (Thinking Mode) | 32768 | Max Chain-of-Thought Length | 81920 |
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)
- Singapore
- Germany (Frankfurt)
- Hong Kong (China)
Input<=32k
Input<=32k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.359 | Per 1M tokens |
Output | 1.434 | Per 1M tokens |
32k<Input<=128k
32k<Input<=128k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.574 | Per 1M tokens |
Output | 2.294 | Per 1M tokens |
128k<Input<=256k
128k<Input<=256k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 1.004 | Per 1M tokens |
Output | 4.014 | Per 1M tokens |
Rate Limits
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- Hong Kong (China)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 600 |
TPM (Tokens Per Minute) | 1,000,000 |
qwen3-max-2025-09-23
The Qwen 3 series Max model has undergone specialized upgrades in agent programming and tool invocation compared to the preview version. The officially released model this time has achieved state-of-the-art (SOTA) performance in its field and is better suited to meet the demands of agents operating in more complex scenarios.This version is a snapshot as of September 23, 2025.
Inference Service Provider
The inference service provider for qwen3-max-2025-09-23 is Alibaba Cloud Model Studio.
Model Capabilities
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
| Capability | Support | Capability | Support |
|---|---|---|---|
Input Modality | Text | Output Modality | Text |
Model Experience | Supported | Function Calling | Supported |
Structured Outputs | Supported | Web Search | Supported |
Prefix Completion | Supported | Context Caching | Unsupported |
Batch Inference | Unsupported | Fine-tuning | Unsupported |
Context Limits
| Parameter | Value | Parameter | Value |
|---|---|---|---|
Max Input Length | 258048 | Max Output Length | 32768 |
Context Window | 262144 |
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)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
Input<=32k
Input<=32k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 0.86 | Per 1M tokens |
Output | 3.44 | Per 1M tokens |
32k<Input<=128k
32k<Input<=128k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 1.434 | Per 1M tokens |
Output | 5.734 | Per 1M tokens |
128k<Input<=256k
128k<Input<=256k
| Billing Item | Price (USD) | Unit |
|---|---|---|
Input | 2.15 | Per 1M tokens |
Output | 8.601 | Per 1M tokens |
Rate Limits
- China (Beijing)
- Singapore
- Germany (Frankfurt)
- US (Virginia)
| Parameter | Value |
|---|---|
RPM (Requests Per Minute) | 60 |
TPM (Tokens Per Minute) | 100,000 |