← All models

Qwen3-Coder-Flash

Published Publisher Alibabaqwen3-coder-flash-2025-07-28

Model Positioning and Release Dates

Qwen3-Coder-Flash is a code generation model based on Qwen3 that inherits the coding agent capabilities of Qwen3-Coder-Plus. This API ID corresponds to the July 28, 2025 snapshot; the Alibaba Cloud international site announcement lists it in the July 29, 2025 release record, while the domestic site announcement lists it in the August 5, 2025 availability record. (docs.modelstudio.console.alibabacloud.com)

Capabilities and Uses

The model accepts text input and produces text output. According to the official documentation, it supports multi-turn tool interactions and is optimized for code repository-level understanding and tool-calling stability; it is suitable for code generation, repository-level code understanding, and coding agent tasks. The availability of specific API features depends on the deployment region. (docs.modelstudio.console.alibabacloud.com)

Limitations

  • The officially listed context window is 1,000,000 tokens, with a maximum input of 997,952 tokens and a maximum output of 65,536 tokens. (docs.modelstudio.console.alibabacloud.com)
  • The snapshot version's feature table shows that Function Calling is supported in the Beijing region but not in the Singapore, Frankfurt, or Virginia regions; support for prefix continuation is the reverse. All regions are listed as not supporting structured output, web search, context caching, batch inference, or model tuning. Check the API documentation for the actual region when using the model. (docs.modelstudio.console.alibabacloud.com)
  • Generated code still requires testing and human review, especially in security-sensitive scenarios.

Official Pricing

The following are Alibaba Cloud Model Studio's standard pay-as-you-go prices as queried, applicable to the specified snapshot version. Pricing tiers are based on the number of input tokens per request; input and output are billed separately, in units of per million tokens. Caching, batch processing, and limited-time promotions are excluded. Domestic CNY prices and international USD prices apply under different account and region conditions and should not be treated as the same package with directly convertible prices. Query date: November 18, 2026. (help.aliyun.com)

Input token tier per request Beijing: input / output (CNY) Singapore international deployment: input / output (USD) Virginia global deployment: input / output (USD)
0 < input ≤ 32K 1 / 4 0.3 / 1.5 0.144 / 0.574
32K < input ≤ 128K 1.5 / 6 0.5 / 2.5 0.216 / 0.861
128K < input ≤ 256K 2.5 / 10 0.8 / 4 0.359 / 1.434
256K < input ≤ 1M 5 / 25 1.6 / 9.6 0.717 / 3.584

The official international pricing table lists the Frankfurt global deployment prices as the same as the Virginia prices in the table. Prices and service support may change; refer to the current information for the relevant site, account, and region. (alibabacloud.com)

Official Sources