QwenLM/Qwen3.6

QwenLM/Qwen3.6

Description: Qwen3.6 is the large language model series developed by Qwen team, Alibaba Group.

License: Apache-2.0

Stars: 3542

Forks: 236

Open issues: 19

Created: 2025-09-11T05:32:39Z

Pushed: 2026-06-03T12:14:19Z

Default branch: main

Fork: no

Archived: no

README:

Qwen3.6

💜 Qwen Studio | 🤗 Hugging Face | 🤖 ModelScope | 📑 Paper | 📖 Documentation | 💬 WeChat (微信) | 🫨 Discord

Welcome to the GitHub repository of Qwen3.6 (& Qwen3.5). Here, you can find official information about Qwen3.6 (User Guide, coming soon), post your questions (Issues), and share your ideas with the community (Discussions).

Introduction

Qwen3.6

Qwen3.6 is the latest addition to the Qwen model family. Building upon the fundamental breakthroughs of Qwen3.5, this release prioritizes stability and real-world utility. It offers developers a more intuitive, responsive, and genuinely productive coding experience, shaped by direct community feedback. This update delivers substantial upgrades, particularly in:

• Agentic Coding: The model now handles front-end workflows and repository-level reasoning with greater fluency and precision.
• Thinking Preservation: A new feature retains thinking context across conversation history, streamlining iterative development and reducing overhead.

Qwen3.5

Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance. Qwen3.5 represents a significant leap forward, integrating breakthroughs in multimodal learning, architectural efficiency, reinforcement learning scale, and global accessibility to empower developers and enterprises with unprecedented capability and efficiency.

Qwen3.5 features the following enhancement:

• Unified Vision-Language Foundation: Early fusion training on trillions of multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.

• Efficient Hybrid Architecture: Gated Delta Networks combined with sparse Mixture-of-Experts deliver high-throughput inference with minimal latency and cost overhead.

• Scalable RL Generalization: Reinforcement learning scaled across million-agent environments with progressively complex task distributions for robust real-world adaptability.

• Global Linguistic Coverage: Expanded support to 201 languages and dialects, enabling inclusive, worldwide deployment with nuanced cultural and regional understanding.

• Next-Generation Training Infrastructure: Near-100% multimodal training efficiency compared to text-only training and asynchronous RL frameworks supporting massive-scale agent scaffolds and environment orchestration.

News

• 2026-04-22: Qwen3.6-27B is now availabe on Hugging Face Hub and ModelScope. Read more on our release blog!
• 2026-04-16: Qwen3.6-35B-A3B is now availabe on Hugging Face Hub and ModelScope. Read more on our release blog!
• 2026-03-02: Qwen3.5-9B, Qwen3.5-4B, Qwen3.5-2B, and Qwen3.5-0.8B are now available on Hugging Face Hub and ModelScope!
• 2026-02-24: Qwen3.5-122B-A10B, Qwen3.5-35B-A3B, and Qwen3.5-27B are released. Check out the model cards on Hugging Face Hub or ModelScope for more information!
• 2026-02-16: We release Qwen3.5. The first release includes a 397B-A17B MoE model. Read more on our release blog. More sizes are coming & Happy Chinese New Year!
• 2025-09-11: We release Qwen3-Next-80B-A3B, an ultra-sparse mixture-of-experts model with hybrid attention architecture, designed for extreme efficiency. Read more on our blog.

Models

The official model weights are released on:

• 🤗Hugging Face Hub: Most LLM frameworks and applications support downloading model files from Hugging Face Hub automatically by specifying the model ID, e.g., Qwen/Qwen3.6-35B-A3B and Qwen/Qwen3.5-397B-A17B.

You can also download model files manually using huggingface download or git clone. Please follow the instructions on the model page.

• 🤖ModelScope: For users unable to access Hugging Face Hub, we strongly recommend using ModelScope.

For supported frameworks, you can download from ModelScope by setting environment variables, such as SGLANG_USE_MODELSCOPE=true or VLLM_USE_MODELSCOPE=true. You can also download model files manually using modelscope download or git clone. Please follow the instructions on the model page.

Benchmarks

Qwen3.6 Open Models

!Qwen3.6-27B Benchmark Results

!Qwen3.6-35B-A3B Benchmark Results

For detailed results, please check out the Qwen3.6-35B-A3B blog and the Qwen3.6-27B blog.

Qwen3.5 Open Models

!Qwen3.5-397B-A17B Benchmark Results

!Qwen3.5-122B-A10B, Qwen3.5-35B-A3B, and Qwen3.5-27B Benchmark Results

!Qwen3.5-9B and Qwen3.5-4B Benchmark Results

For detailed results, please check out the Qwen3.5 blog.

Quickstart

To learn more about Qwen3.6, feel free to read our documentation (coming soon).

Official

You can try Qwen3.6 on our official sites and enjoy the native experience with extra features, such as deep research, web dev, and adaptive tool use.

Qwen Studio

For users who simply would…