Qwen: Qwen3 235B A22B Thinking 2507

qwen/qwen3-235b-a22b-thinking-2507

Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144 tokens of context. This "thinking-only" variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs (up to 81,920 tokens) in challenging domains. The model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases.

Model specifications

Context
262,144 tokens
Max output
32,768 tokens
Input price
$0.28 / 1M tokens
Output price
$2.3 / 1M tokens
Released
2026-02-05

Capabilities

  • Streaming
  • Playground

Provider pricing, discounts and data privacy

Compare effective provider prices, published discounts, regions, retention policies, training use, compliance, and privacy links by service tier.

Standard service tier

2 available providers · tier input average $0.255 / 1M tokens · tier output average $2.3 / 1M tokens

Alibaba Cloud Int.(SG)

Tier: Standard · Region: SG · Quantization: int8

Pricing
Input
$0.23 / 1M tokens
Output
$2.3 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
SG
Zero data retention
No
Data retention
30-day retention
Used for training
No
Data collection
Moderated
Yes
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
Yes

AtlasCloud

Tier: Standard · Region: US

Pricing
Input
$0.28 / 1M tokens
Output
$2.3 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
US
Zero data retention
No
Data retention
7-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
No
HIPAA compliant
No
SOC 2 certified
No
BYOK supported
No

Frequently asked questions

What is Qwen: Qwen3 235B A22B Thinking 2507?
Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144 tokens of context. This "thinking-only" variant enhances structured logical reasoning, mathematics, science, and long-form generation, showing strong benchmark performance across AIME, SuperGPQA, LiveCodeBench, and MMLU-Redux. It enforces a special reasoning mode (</think>) and is designed for high-token outputs (up to 81,920 tokens) in challenging domains. The model is instruction-tuned and excels at step-by-step reasoning, tool use, agentic workflows, and multilingual tasks. This release represents the most capable open-source variant in the Qwen3-235B series, surpassing many closed models in structured reasoning use cases.
How much does Qwen: Qwen3 235B A22B Thinking 2507 cost?
Input costs start at $0.28 / 1M tokens and output costs start at $2.3 / 1M tokens. Provider-level prices vary by service tier.
What is the context length of Qwen: Qwen3 235B A22B Thinking 2507?
Qwen: Qwen3 235B A22B Thinking 2507 supports a 262,144 token context window and up to 32,768 output tokens.
What capabilities does Qwen: Qwen3 235B A22B Thinking 2507 support?
Qwen: Qwen3 235B A22B Thinking 2507 supports Streaming, Playground.
Which providers offer Qwen: Qwen3 235B A22B Thinking 2507?
Qwen: Qwen3 235B A22B Thinking 2507 is available from Alibaba Cloud Int.(SG), AtlasCloud.
How do providers handle data privacy for Qwen: Qwen3 235B A22B Thinking 2507?
2 of 2 providers report that customer data is not used for training, and 0 offer zero-data-retention routing. Retention, compliance, and privacy-policy links are listed per provider.

Browse all AI models