Qwen/Qwen3.6-27B

Entry date: 2026-09-08

We are excited to announce that we are upgrading our model lineup by introducing Qwen/Qwen3.8-27B, which will serve as the successor to our current Qwen offerings.

As the successor of Qwen3.6 27B the newer Qwen3.8 27B generally improves quality across all fields of application, while maintaining the focus on coding tasks. It was trained on a broad mix of publicly available data and post-trained for code generation, debugging, and technical reasoning, while also supporting general assistant tasks. With a context window of 262K tokens, it can handle large codebases and complex multi-step workflows.

As part of this transition, we are officially deprecating the following model:

• Qwen/Qwen3.6-27B

We kindly ask all customers to migrate their workloads to the new model Qwen/Qwen3.8-27B before December 08, 2026.

Read from the provider's text and verified a second time against the quoted lines.

Entry date: 2026-05-22

We are excited to announce the release of Qwen3.6 27B, a coding-focused language model from the Qwen team, to our shared LLM model portfolio.

Qwen3.6 27B is post-trained for code generation, debugging, and technical reasoning, while also supporting general assistant tasks. With a context window of 262K tokens, it can handle large codebases and complex multi-step agentic workflows.

Read from the provider's text; the quote is the provider's exact lines.