Tooony133/Qwen-3.6-27B-AstroWolf

VISIONConcurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jul 6, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Qwen3.6-27B is a 27 billion parameter causal language model with a vision encoder, developed by Qwen. This model prioritizes stability and real-world utility, excelling in agentic coding tasks, including frontend workflows and repository-level reasoning. It features a native context length of 262,144 tokens, extensible up to 1,010,000 tokens, and introduces a new option to preserve reasoning context from historical messages for streamlined iterative development.

Loading preview...

Qwen3.6-27B: Enhanced Agentic Coding and Context Preservation

Qwen3.6-27B is a 27 billion parameter causal language model with a vision encoder, developed by Qwen, focusing on stability and real-world utility for developers. It builds upon the Qwen3.5 series with substantial upgrades, particularly in agentic coding and thinking preservation.

Key Capabilities

  • Agentic Coding: Significantly improved handling of frontend workflows and repository-level reasoning, offering greater fluency and precision.
  • Thinking Preservation: A new feature allows retaining reasoning context from historical messages, streamlining iterative development and reducing overhead. This also improves KV cache utilization.
  • Multimodal: Supports text, image, and video inputs, making it versatile for various applications.
  • Extended Context: Natively supports a context length of 262,144 tokens, extensible up to 1,010,000 tokens using RoPE scaling techniques like YaRN.
  • Strong Performance: Achieves competitive scores across various benchmarks, including SWE-bench (77.2% verified), MMLU-Pro (86.2%), and MMMU (82.9%), demonstrating strong capabilities in coding, knowledge, STEM, and vision-language tasks.

Good for

  • Complex Coding Tasks: Ideal for developers requiring advanced agentic coding capabilities, including repository-level understanding and iterative development.
  • Multimodal Applications: Suitable for use cases involving image and video understanding alongside text.
  • Long-Context Processing: Beneficial for applications requiring processing and maintaining context over very long sequences of text or code.
  • Agent Development: Recommended for building robust AI agents, especially with frameworks like Qwen-Agent and Qwen Code, due to its enhanced tool-calling and thinking preservation features.

Finetunes
1 models