DISCOVER & COMPARE
探索 AI 模型 16
为你发现值得关注的模型 基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 待核验
A Swin-B image feature model. Supervisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 109.1 - Image size: 224 x 224 - Architecture: swin_base_patch4_window7_224 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - Swin Transformer: Hierarchical Vision Transformer using Shifted Windows - DINOv2: Learning Robust Visual Features without Supervision - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A CLIP based image feature model. Supervisely pre-trained on animal re-identification datasets. Original model: timm/vit_large_patch14_clip_336.openai_ft_in12k_in1k - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 304.5 - GMACs: 174.7 - Activations (M): 128.2 - Image size: 336 x 336
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A DINOv2 based image feature model. Supervisely pre-trained on animal re-identification datasets. Original model: timm/vit_large_patch14_dinov2.lvd142m - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 304.4 - GMACs: 507.1 - Activations (M): 1058.8 - Image size: 518 x 518
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A Swin-L image feature model. Supervisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 228.6 - Image size: 224 x 224 - Architecture: swin_large_patch4_window7_224 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - Swin Transformer: Hierarchical Vision Transformer using Shifted Windows - DINOv2: Learning Robust Visual Features without Supervision - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A Swin-L image feature model. Superwisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 228.8 - Image size: 384 x 384 - Architecture: swin_large_patch4_window12_384 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - Swin Transformer: Hierarchical Vision Transformer using Shifted Windows - DINOv2: Learning Robust Visual Features without Supervision - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A Swin-S image feature model. Supervisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 49.6 - Image size: 224 x 224 - Architecture: swin_small_patch4_window7_224 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - Swin Transformer: Hierarchical Vision Transformer using Shifted Windows - DINOv2: Learning Robust Visual Features without Supervision - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A Swin-T image feature model. Superwisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Params (M): 28.3 - Image size: 224 x 224 - Architecture: swin_tiny_patch4_window7_224 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - Swin Transformer: Hierarchical Vision Transformer using Shifted Windows - DINOv2: Learning Robust Visual Features without Supervision - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
A EfficientNetB3 based image feature model. Supervisely pre-trained on animal re-identification datasets. - **Model Type:** Animal re-identification / feature backbone - **Model Stats:** - Architecture: EfficientNetB3 - Params (M): 12.2 - GMACs: 1.6 - Activations (M): 21.5 - Image size: 288 x 288 - **Paper:** WildlifeDatasets_An_Open-Source_Toolkit_for_Animal_Re-Identification - **Related Papers:** - EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks - **Pretrain Dataset:** All available re-identification datasets --> https://github.com/WildlifeDatasets/wildlife-datasets
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
--- tags: - image-classification - ecology - fungi - FGVC library_name: DanishFungi license: cc-by-nc-4.0 ---
译文排队中 · 当前显示来源原文
输入 图像→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 图像
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
基础资料已收录 · 缺失资料后续补充
输入 图像→ 输出 待核验
基础档案已收录，详细介绍后续补充。可先查看厂商、分类与来源。
输入 图像→ 输出 待核验
暂未找到匹配模型
可以调整厂商、分类或输入输出条件；新模型正持续收录。
调整月度 Token 用量，直观比较不同模型的预估 API 开支。
使用已采集且单位可比的 USD / 百万 Token 渠道报价。报价为采集时快照；不计缓存优惠、阶梯价、推理额外输出与税费，请以来源的当前报价为准。
API 价格应该怎么比较？+
先区分输入、输出和缓存价格，再使用同一任务和用量估算总费用。图像、音频和视频可能按张、时长或分辨率计费，不能直接与 Token 单价比较。
开放权重，就代表可以免费商用吗？+
不一定。必须逐一核对具体模型版本的许可证、商用限制及使用政策。自己部署还需要考虑显存、推理资源和运维费用。
排行榜第一，就是最适合我的模型吗？+
榜单取决于测试任务、版本与评测方法。建议用实际业务样本比较准确性、时延、稳定性和成本，再选择模型。
