DavidAU/Qwen3.5-27B-Gemini3-Pro-High-Reasoning-Compact-Thinking

VISIONConcurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Mar 1, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Featherless Exclusive Cold

DavidAU/Qwen3.5-27B-Gemini3-Pro-High-Reasoning-Compact-Thinking is a 27 billion parameter multimodal language model fine-tuned from the Qwen 3.5 base model, featuring a 32768 token context length. Developed by DavidAU, this model incorporates Gemini-like reasoning and thinking blocks, optimized for enhanced reasoning capabilities while maintaining strong benchmarks. It supports vision inputs and is designed for complex tasks requiring detailed thought processes.

Loading preview...

Model Overview

DavidAU/Qwen3.5-27B-Gemini3-Pro-High-Reasoning-Compact-Thinking is a 27 billion parameter multimodal model, fine-tuned from the Qwen 3.5 base, with a native context length of 32768 tokens, extensible up to 1,010,000 tokens using YaRN scaling. This model integrates Gemini-like reasoning and thinking blocks, with altered block sizes for compact thinking, aiming to enhance reasoning without compromising the base model's strong performance. It supports both text and vision inputs, with video input capabilities also noted.

Key Capabilities

  • Enhanced Reasoning: Incorporates Gemini-like thinking blocks for improved complex problem-solving.
  • Multimodal Understanding: Processes both text and image inputs, with video understanding capabilities.
  • Long Context Handling: Natively supports 32K tokens, extensible to over 1M tokens with YaRN.
  • Strong Benchmarks: Maintains competitive performance across various language and vision-language benchmarks, including MMLU, IFEval, MMMU, and MathVision.
  • Agentic Usage: Excels in tool calling and is recommended for use with Qwen-Agent and Qwen Code for building agent applications.

Good for

  • Applications requiring advanced reasoning and problem-solving.
  • Multimodal tasks involving text, images, and potentially video analysis.
  • Use cases demanding long context understanding and generation.
  • Developing AI agents that leverage tool-use capabilities.