InternVL3_5-2B-Flash

OpenGVLab/InternVL3_5-2B-Flash

PublisherOpenGVLab

Download access requirementsMirror available

InternVL3.5 5-2B-Flash InternVL3.5是一个开源多模态模型系列，旨在提升多模态任务的通用性、推理能力和推理效率。该模型支持图像与文本的联合理解与推理，具备GUI交互和具身代理等新能力。 InternVL3.5采用级联强化学习框架，通过离线强化学习实现稳定收敛，再通过在线强化学习进行精细对齐，显著提升下游推理任务性能。模型引入视觉分辨率路由器，可动态调整视觉token分辨率而不影响性能。解耦视觉-语言部署策略将视觉编码器和语言模型部署在不同GPU上

Vision & LanguageEmbedding & RetrievalText Chat

View all download instructions Mirror repository documentation and license ↗

Model details

模型介绍

中文整理

InternVL3.5 5-2B-Flash

Model use case

InternVL3.5是一个开源多模态模型系列，旨在提升多模态任务的通用性、推理能力和推理效率。该模型支持图像与文本的联合理解与推理，具备GUI交互和具身代理等新能力。

主要能力

InternVL3.5采用级联强化学习框架，通过离线强化学习实现稳定收敛，再通过在线强化学习进行精细对齐，显著提升下游推理任务性能。模型引入视觉分辨率路由器，可动态调整视觉token分辨率而不影响性能。解耦视觉-语言部署策略将视觉编码器和语言模型部署在不同GPU上，有效平衡计算负载。InternVL3.5-Flash版本通过视觉一致性学习，将视觉token压缩至64个，在保持接近100%性能的同时减少50%的视觉token数量。模型支持深度思考模式，可进行逐步推理；支持并行思考策略，从多个候选答案中选择最优响应。

输入输出

输入：图像和文本的组合序列

输出：文本响应运行要求

模型提供多种参数规模版本，包括1B、2B、4B、8B、14B、38B等版本。启用思考模式需将系统提示设置为R1 SYSTEM PROMPT。建议设置do_sample=True和temperature=0.6以减少不必要的重复。上下文窗口设置为32K token以适应长上下文信息。

基本用法

使用VLMEvalkit进行评估。如不确定选择哪个版本，请选择不带后缀的版本，该版本已完成完整训练流程。

限制

除非另有说明，论文中报告的实验结果未应用测试时扩展。测试时扩展目前仅应用于推理基准测试，因为模型已展现出较强的感知和理解能力，启用测试时扩展未带来显著提升。

License

开源模型，模型权重和代码已公开发布。

优先采用原作者提供的中文模型卡片；仅有英文说明时由闲社AI翻译整理。重要参数、许可证和商用范围请以原文为准。

你将获得什么

完整仓库方式包含该固定版本中的权重、配置、分词器及说明文件；指定版本方式只下载所选文件或整组分片。多模态模型的投影文件、配置等依赖，请按作者说明配齐。

完整仓库可能包含多个权重格式，因此需要的磁盘空间可能大于单一格式的模型大小。下载文件不等于完成模型部署。

使用前须知

运行环境

运行模型需要的设备、工具与依赖，以原作者说明为准。

授权范围

是否允许商用、修改和分发，请查看模型许可证。

闲社提供资料与镜像下载方法，不代表已运行模型或完成安全审计。下载的代码应先检查，再执行。

Problem with the information or download?

Pinned version · Multiple download options

InternVL3_5-2B-Flash model download

Download the full repository or choose a version or file.

Download tools · More than GGUF

Use the Hugging Face download tool to obtain OpenGVLab/InternVL3_5-2B-Flash. Download Safetensors, GGUF, ONNX, configuration and other repository files. Downloading does not mean the tool can run this model.

Pinned-version file information is being completed. Existing downloads remain available, or load the file manifest below.

Choose a mirror or official source, your operating system and download scope below, then copy the command. Install the download tool first if needed.

Pinned version: 43902d0577f0bb49386d60a0c31da4eb6e293462. Downloads files only; does not start the model.

Choose a download source

已根据访问地区推荐 Hugging Face 官方线路，你可以手动切换。

Choose what to download

Need just one GGUF precision? Load the file manifest and choose a version. Direct file links do not require Python; the command below can download multiple shards together.

Quantization / file

This reads the pinned version's file manifest; no model weights are downloaded.

Choose your operating system

For command-line downloads, follow these steps to get the full repository or selected files. Direct file links do not need these steps.

Open a terminal window

Create and open a folder for the model. In the folder's address bar, enter powershell and press Enter.

If Python is not installed, get it from the official Python website. In the terminal, enter python --version; continue when a version number appears.

First time? Install the download tool

Copy this command, paste it into the terminal, press Enter and wait for installation to finish.

python -m pip install --upgrade huggingface_hub

Already installed? Continue to step 3.

Run the download command

Select the button below, paste into the same terminal and press Enter. The tool downloads all files included in the current command.

Copying does not start a download. Paste and run the command on your device.

Where are the downloaded files?

Open the folder you chose. Look for the models folder. When progress stops and the terminal returns to its prompt without errors, check that all files are present. 'Copied' does not mean 'downloaded'.

No response or an error?

python not found: Install Python and reopen the terminal. On Windows, follow the installer's instructions to enable command-line access.

hf not found: Confirm that step 2 succeeded, then reopen the terminal. If needed, add Python's Scripts folder to your user PATH. Do not repeatedly download scripts.

Network interrupted: Stay in the same folder and run the same command again. The tool checks existing files.

Permission or disk error: Choose a writable folder with enough free space. Do not disable system security protection.

View the full download command

$env:HF_ENDPOINT = "https://huggingface.co"
$env:HF_HUB_DISABLE_IMPLICIT_TOKEN = "1"
$env:HF_HUB_DISABLE_XET = "1"
hf download "OpenGVLab/InternVL3_5-2B-Flash" --revision "43902d0577f0bb49386d60a0c31da4eb6e293462" --local-dir "./models/OpenGVLab--InternVL3_5-2B-Flash"

Files transfer directly from the source you choose, not through Xianshe servers.

The version is pinned. Repeating the same command checks existing files. Change the source manually if it is slow or unavailable.

Public models do not require submitting keys to Xianshe. Restricted models still require the author's authorization. On Linux / macOS, use python3 instead of python when appropriate.