TGTGInsighttelegram intelligenceLIVE / telegram public index
← GitHub Trends

TGINSIGHT SIMILAR POSTS

Find similar content

Source channel @githubtrending · Post #14627 · Apr 24

#jupyter_notebook DINOv2 is a powerful AI model from Meta AI that learns to understand images without needing labeled data, using self-supervised learning. It was trained on 142 million images and creates strong visual features that work well for many tasks like image classification, depth estimation, and segmentation without extra fine-tuning. You can use its pretrained models easily with simple classifiers, saving time and effort. DINOv2 is efficient, scalable, and performs better than many other models, making it great for building versatile computer vision applications quickly and accurately. It’s open-source and ready to use with PyTorch. https://github.com/facebookresearch/dinov2

Results

1 similar post found

Search: #apachelicense

当前筛选 #apachelicense清除筛选
AI一线|ShareCentre

@ShareCentre · Post #7154 · 04/03/2026, 02:34 AM

Google 发布 Gemma 4 开源模型:Apache 2.0 许可,31B 登顶 Arena AI 开源第三 Google DeepMind 于 4 月 2 日正式发布 Gemma 4,迄今最强大的开源模型家族。基于与 Gemini 3 相同的技术构建,专为高级推理和 Agent 工作流打造。全系列首次采用 Apache 2.0 开源许可,完全商用自由。 自初代发布以来,Gemma 已被下载超过 4 亿次,社区衍生出 10 万+变体。 ⚙️ 四款模型 - E2B(Effective 2B):手机/IoT 边缘设备,128K 上下文,支持文本+图像+视频+音频 - E4B(Effective 4B):手机/IoT 边缘设备,128K 上下文,支持文本+图像+视频+音频 - 26B MoE(混合专家):推理时仅激活 3.8B 参数,256K 上下文,Arena AI 开源第 6 - 31B Dense(稠密):最大化质量,256K 上下文,Arena AI 开源第 3,超越 20 倍参数量模型 🔑 核心能力 - 高级推理:多步规划与深度逻辑 - Agent 工作流:原生函数调用、结构化 JSON 输出、系统指令 - 代码生成:高质量离线代码,本地 AI 编程助手 - 多模态:全系列处理图像和视频,E2B/E4B 额外支持音频 - 140+ 语言原生训练 - Apache 2.0 许可:完全商用自由 📎 相关链接 官方博客:https://blog.google/innovation-and-ai/technology/developers-tools/gemma-4/ Google AI Studio:https://aistudio.google.com Hugging Face:https://huggingface.co/collections/google/gemma-4 Kaggle:https://www.kaggle.com/models?query=gemma-4&publisher=google #Google#Gemma4#开源模型#ApacheLicense#AI