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。
探索 AI 模型 85001
This model was converted to GGUF format from `Qwen/Qwen2.5-Coder-7B-Instruct` using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model. Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well. Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
译文排队中 · 当前显示来源原文
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
nemotron-3.5-asr-streaming-litert
konradjr007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
beans-deit-tiny
Konradowsky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。
relighting-kontext-dev-lora-v3
kontext-community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
sbert_punc_case_ru
kontur-aiSbertPuncCase - модель восстановления пунктуации и регистра для русского языка. Модель способна расставлять точки, запятые и знаки вопроса; определять регистр - слово в нижнем регистре, слово с первой буквой в верхнем регистре, слово в верхнем регистре. Модель разработана для восстановления текста после распознавания речи, поэтому работает со строками в нижнем регистре. В основу модели легла sbert_large_nlu_ru. В качестве обучающих данных использованы текстовые расшифровки интервью. 1. Текст переводится в нижний регистр и разбивается на слова. 2. Слова разделяются на токены. 3. Модель (по аналогии с задачей NER) предсказывает класс для каждого токена. Классификация на 12 классов: 3+1 знака препинания * 3 варианта регистра. 4. Функция декодировки восстанавливает текст соответственно предсказанным классам.
译文排队中 · 当前显示来源原文
AnimeBoysXL-v3.0
Koolchh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
MiniMax_H3_int4_W4A8_ConvRot_Pruned
koongrizzlyhttps://github.com/Koongrizzly/MiniMax_H3_Standalone_app (also allows loading some int8 models) For questions about the app or repo : https://discord.gg/kWxAJwjSY This repository contains a collection of MiniMax H3 model components intended for lower-memory local inference. MiniMax H3 is an omni-modal video generation model capable of generating synchronized video and stereo audio. The official release supports text-to-audio-video, first/last-frame-to-audio-video, and multimodal reference-to-audio-video generation, depending on the selected model variant.
译文排队中 · 当前显示来源原文
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
Qwen2.5-3B-Instruct-heretic
kor-448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
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
Mi:dm K 2.5 Pro
Korea Telecom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
Mi:dm K 2.5 Pro Preview
Korea Telecom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
ms-eff-gcvit-deepfake-b0-celeb-df-v2
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. A large-scale challenging dataset for deepfake forensics [[Paper]](https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Celeb-DF_A_Large-Scale_Challenging_Dataset_for_DeepFake_Forensics_CVPR_2020_paper.pdf) [[Download]](https://github.com/yuezunli/celeb-deepfakeforensics/tree/master), featuring **590 YouTube celebrity videos** with diverse ages, ethnic groups, and genders.
译文排队中 · 当前显示来源原文
ms-eff-gcvit-deepfake-b0-ff-plus-plus
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. Learning to detect manipulated facial images [[Paper]](https://arxiv.org/abs/1901.08971) [[Download]](https://github.com/ondyari/FaceForensics), featuring **1,000 original YouTube videos** manipulated by 5 face forgery methods.
译文排队中 · 当前显示来源原文
ms-eff-gcvit-deepfake-b0-kodf
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. While traditional Vision Transformers (ViTs) utilize a Linear Projection for patch embedding, our proposed model adopts a CNN-based Patch Embedding module incorporating MBConvBlocks.
译文排队中 · 当前显示来源原文
ms-eff-gcvit-deepfake-b5-celeb-df-v2
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. A large-scale challenging dataset for deepfake forensics [[Paper]](https://openaccess.thecvf.com/content_CVPR_2020/papers/Li_Celeb-DF_A_Large-Scale_Challenging_Dataset_for_DeepFake_Forensics_CVPR_2020_paper.pdf) [[Download]](https://github.com/yuezunli/celeb-deepfakeforensics/tree/master), featuring **590 YouTube celebrity videos** with diverse ages, ethnic groups, and genders.
译文排队中 · 当前显示来源原文
ms-eff-gcvit-deepfake-b5-ff-plus-plus
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. Learning to detect manipulated facial images [[Paper]](https://arxiv.org/abs/1901.08971) [[Download]](https://github.com/ondyari/FaceForensics), featuring **1,000 original YouTube videos** manipulated by 5 face forgery methods.
译文排队中 · 当前显示来源原文
ms-eff-gcvit-deepfake-b5-kodf
KoreaPeterMulti-Scale Efficient Global Context Vision Transformer (MS-EffGCViT) is a hybrid CNN-ViT architecture for **deepfake detection**. It fuses CNN-driven spatial inductive bias with hierarchical global-context attention to catch both *local* artifacts (textures, blending seams) and *global* artifacts (lighting, structural inconsistency). A single architecture ships in two sizes and three domain-tuned checkpoints, working on both **static images** and **video** at the **frame level**. While traditional Vision Transformers (ViTs) utilize a Linear Projection for patch embedding, our proposed model adopts a CNN-based Patch Embedding module incorporating MBConvBlocks.
译文排队中 · 当前显示来源原文
en_core_med7_lg
kormilitzin`en_core_med7_lg` is a non-transformer, large-vector spaCy pipeline for medication-related named entity recognition in English clinical text. Unlike `en_core_med7_trf`, this model does not use a transformer backend and does not require `spacy-transformers`, `transformers`, `tokenizers`, or `torch` at runtime. Kormilitzin, A., Vaci, N., Liu, Q., & Nevado-Holgado, A. (2021). Med7: A transferable clinical natural language processing model for electronic health records. *Artificial Intelligence in Medicine*, 118, 102086. https://doi.org/10.1016/j.artmed.2021.102086
译文排队中 · 当前显示来源原文
KORMo-10B-base
KORMo-Team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
KORMo-10B-sft
KORMo-Team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
ConGen-WangchanBERT-Small
kornwtp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
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。
好模型,也要算好成本。
调整月度 Token 用量,直观比较不同模型的预估 API 开支。
仅比较同一渠道的基础文本 Token 报价
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。
读懂模型,再做选择
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。