shatu/Qwen3.5-4B-Reasoning-Fix

VISIONConcurrent Unit Cost:1Model Size:4.5BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Jun 26, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

Qwen3.5-4B-Reasoning-Fix is a 4.5 billion parameter causal language model with a vision encoder developed by Qwen. This model integrates multimodal learning and an efficient hybrid architecture, featuring Gated Delta Networks and sparse Mixture-of-Experts. It is optimized for reasoning, coding, and visual understanding benchmarks, supporting a native context length of 262,144 tokens and extensible up to 1,010,000 tokens for complex tasks.

Loading preview...

Overview

Qwen3.5-4B is a 4.5 billion parameter multimodal causal language model from the Qwen family, designed for exceptional utility and performance. It features a unified vision-language foundation through early fusion training, achieving strong performance across reasoning, coding, agent, and visual understanding benchmarks. The model utilizes an efficient hybrid architecture combining Gated Delta Networks with sparse Mixture-of-Experts for high-throughput inference.

Key Capabilities

  • Multimodal Learning: Processes both text and visual inputs, including images and videos, with strong performance on STEM, puzzle, and general VQA tasks.
  • Extended Context: Natively supports 262,144 tokens, extensible up to 1,010,000 tokens using YaRN scaling for ultra-long texts.
  • Multilingual Support: Expanded linguistic coverage to 201 languages and dialects.
  • Agentic Functionality: Excels in tool calling, with specific optimizations for Qwen-Agent and Qwen Code frameworks.
  • Reasoning and Coding: Demonstrates strong performance in reasoning and coding benchmarks, including HMMT and LiveCodeBench.

When to Use This Model

This model is particularly well-suited for:

  • Applications requiring advanced multimodal understanding, such as image and video analysis combined with natural language processing.
  • Tasks demanding long context processing, like summarizing extensive documents or complex codebases.
  • Developing AI agents that interact with tools or operate in complex environments.
  • Scenarios needing strong reasoning capabilities across various domains, including STEM and general knowledge.
  • Global applications benefiting from broad linguistic coverage.