Fluxmire/Qwen-3.6-27B-NEXT-v4

VISIONConcurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 30, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The Qwen3.6-27B is a 27 billion parameter causal language model with a vision encoder, developed by Qwen. It is designed for agentic coding, offering enhanced fluency and precision in frontend workflows and repository-level reasoning. This model also features thinking preservation, allowing it to retain reasoning context from historical messages for streamlined iterative development, and supports a native context length of 262,144 tokens, extensible up to 1,010,000 tokens.

Loading preview...

Qwen3.6-27B: An Agentic Coding and Multimodal Powerhouse

Qwen3.6-27B is a 27 billion parameter causal language model with an integrated vision encoder, developed by Qwen. This model builds upon the Qwen3.5 series, focusing on stability and real-world utility, particularly for developers.

Key Capabilities & Differentiators

  • Agentic Coding: Significantly upgraded to handle frontend workflows and repository-level reasoning with improved fluency and precision. This includes strong performance on benchmarks like SWE-bench Verified (77.2), SWE-bench Pro (53.5), and Terminal-Bench 2.0 (59.3).
  • Thinking Preservation: Introduces a novel feature to retain reasoning context from historical messages, which streamlines iterative development, reduces overhead, and enhances decision consistency in agent scenarios.
  • Multimodal Understanding: As a vision-capable model, it excels in tasks requiring image and video input, demonstrated by strong scores in MMMU (82.9), MathVista mini (87.4), and VideoMMMU (84.4).
  • Extended Context Length: Natively supports a context length of 262,144 tokens, which can be extended up to 1,010,000 tokens using techniques like YaRN, making it suitable for ultra-long text processing.
  • Robust Performance: Achieves competitive results across various benchmarks, including MMLU-Pro (86.2) for knowledge and AIME26 (94.1) for STEM & Reasoning, often outperforming or matching larger models in its class.

Ideal Use Cases

  • Code Generation & Debugging: Developers can leverage its agentic coding capabilities for complex programming tasks, including frontend development and repository-level code analysis.
  • AI Agents: The thinking preservation feature makes it highly suitable for building sophisticated AI agents that require consistent reasoning across multiple turns.
  • Multimodal Applications: Excellent for applications involving image and video analysis, such as visual question answering, document understanding, and spatial intelligence tasks.
  • Long-Context Processing: Ideal for tasks requiring the processing and understanding of very long documents, codebases, or conversational histories.