Qwen3.5 2B

提供商Qwen
分类image-text-to-text
许可证apache-2.0
下载量1.1M
星标43

简介

Qwen3.5 2B 是一款轻量级但能力强悍的多模态模型，主打在端侧设备实现高效的图文理解。它不仅能处理纯文本任务，还能精准识别并分析图像内容，将视觉感知与语言生成深度结合。对于开发者而言，2B 的参数规模意味着极低的部署门槛和推理成本，非常适合集成到手机 App、边缘计算设备或作为复杂工作流中的预处理模块。相比于超大规模模型，它在保证响应速度的同时，在中文语境的理解上依然保持了 Qwen 系列的高水准，是追求性价比和实时性的理想选择。

核心亮点

  • 轻量化端侧部署，极低推理延迟与资源占用
  • 强大的图文跨模态理解，支持复杂视觉分析
  • 中文语境优化，适配多种实际应用场景
  • Apache-2.0 开源协议，商业集成灵活便捷

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("Qwen/Qwen3.5-2B")
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-2B")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download Qwen/Qwen3.5-2B

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download Qwen/Qwen3.5-2B config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('Qwen/Qwen3.5-2B')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/Qwen/Qwen3.5-2B

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/Qwen/Qwen3.5-2B

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('Qwen/Qwen3.5-2B')
tokenizer = AutoTokenizer.from_pretrained('Qwen/Qwen3.5-2B')

模型下载

我们推荐使用命令行或者 ModelScope SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 ModelScope：

操作指引 
pip install modelscope

命令行下载

下载完整模型库

下载完整模型库 
modelscope download --model Qwen/Qwen3.5-2B

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例）

下载单个文件到指定本地文件夹（以下载 README.md 到当前路径下 dir 目录为例） 
modelscope download --model Qwen/Qwen3.5-2B README.md --local_dir ./dir

更多更丰富的命令行下载选项，可参见具体文档

SDK 下载

SDK 下载 
# 模型下载
from modelscope import snapshot_download
model_dir = snapshot_download('Qwen/Qwen3.5-2B')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://www.modelscope.cn/Qwen/Qwen3.5-2B.git

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://www.modelscope.cn/Qwen/Qwen3.5-2B.git

ModelScope 模型页直接下载模型文件；无需将模型文件放在本站服务器。

Notebook 快速开发

下载并安装 ModelScope library

下载并安装 ModelScope library 
pip install "modelscope[audio,cv,nlp,multi-modal,science]" -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html

模型加载和推理

模型加载和推理 
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks

p = pipeline('text-generation', 'Qwen/Qwen3.5-2B')

完整文档

来源: HuggingFace

---
library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.5-2B/blob/main/LICENSE
pipeline_tag: image-text-to-text
base_model:

  • Qwen/Qwen3.5-2B-Base

---

Qwen3.5-2B

<img width="400px" src="https://qianwen-res.oss-accelerate.aliyuncs.com/logo_qwen3.5.png">

![Qwen Chat](https://chat.qwen.ai)

> [!Note]
> This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format.
>
> These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc.
>
> In light of its parameter scale, the intended use cases are prototyping, task-specific fine-tuning, and other research or development purposes.

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 Highlights

Qwen3.5 features the following enhancement:

  • Unified Vision-Language Foundation: Early fusion training on 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.

For more details, please refer to our blog post Qwen3.5.

Model Overview

  • Type: Causal Language Model with Vision Encoder
  • Training Stage: Pre-training & Post-training
  • Language Model
- Number of Parameters: 2B - Hidden Dimension: 2048 - Token Embedding: 248320 (Padded) - Number of Layers: 24 - Hidden Layout: 6 × (3 × (Gated DeltaNet → FFN) → 1 × (Gated Attention → FFN)) - Gated DeltaNet: - Number of Linear Attention Heads: 16 for V and 16 for QK - Head Dimension: 128 - Gated Attention: - Number of Attention Heads: 8 for Q and 2 for KV - Head Dimension: 256 - Rotary Position Embedding Dimension: 64 - Feed Forward Network: - Intermediate Dimension: 6144 - LM Output: 248320 (Tied to token embedding) - MTP: trained with multi-steps
  • Context Length: 262,144 natively

Benchmark Results

Language

<div style="font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif;max-width:1000px;margin:0 auto;padding:16px 0">
<table style="border-collapse:collapse;font-size:13px">
<thead><tr>
<th style="padding:10px 7px;text-align:left;font-weight:600;border-bottom:2px solid #7c3aed;color:#7c3aed"></th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #7c3aed;color:#7c3aed;font-size: 14px;">Qwen3-4B-2507</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #7c3aed;color:#7c3aed;font-size: 14px;">Qwen3-1.7B</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #7c3aed;color:#7c3aed;font-size: 14px;">Qwen3.5-2B</th><th style="padding:10px 7px;text-align:center;font-weight:500;border-bottom:2px solid #7c3aed;color:#7c3aed;font-size: 14px;">Qwen3.5-0.8B</th></tr></thead>
<tbody>
<tr><td colspan="5" style="padding:8px 12px;font-weight:600;color:#7c3aed;border-bottom:1px solid rgba(124, 58, 237, 0.2);background:rgba(124, 58, 237, 0.1)">Instruct (Non-Thinking) Mode</td></tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">MMLU-Pro</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">69.6</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">40.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">55.3</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">29.7</td>
</tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">MMLU-Redux</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">84.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">64.4</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">69.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">48.5</td>
</tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">C-Eval</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">80.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">61.0</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">65.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">46.4</td>
</tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">SuperGPQA</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">42.8</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">21.0</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">30.4</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">16.9</td>
</tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">IFEval</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">83.4</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">68.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">61.2</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">52.1</td>
</tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">MMMLU</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">64.9</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">46.7</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">56.9</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">34.1</td>
</tr>
<tr><td colspan="5" style="padding:8px 12px;font-weight:600;color:#7c3aed;border-bottom:1px solid rgba(124, 58, 237, 0.2);background:rgba(124, 58, 237, 0.1)">Knowledge & STEM (Thinking)</td></tr>
<tr>
<td style="padding:7px 7px;padding-left:20px;border-bottom:1px solid rgba(128, 128, 128, 0.15);">MMLU-Pro</td>
<td style="padding:7px 7px;text-align:center;border-bottom:1px solid rgba(128, 128, 128, 0.15)">74.0</td>
<td

查看官方来源