{ "updatedAt": "2026-05-23T05:38:23.557Z", "date": "2026-05-23", "count": 60, "items": [ { "id": "SulphurAI/Sulphur-2-base", "name": "Sulphur-2-base", "author": "SulphurAI", "pipelineTag": "text-to-video", "pipelineZh": "文生视频", "library": "diffusers", "likes": 1274, "downloads": 1249582, "trendingScore": 243, "createdAt": "2026-05-03T00:33:24.000Z", "lastModified": "2026-05-22T01:12:13.000Z", "tags": [ "text-to-video" ], "source": "HF Trending", "url": "https://huggingface.co/SulphurAI/Sulphur-2-base", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/SulphurAI/Sulphur-2-base.png", "summary": "Sulphur-2-base 是一款基于 LTX 2.3 的文本到视频生成模型支持 t2v 和 i2v 任务具备强大的提示增强能力。", "title_i18n": { "en": "Sulphur-2-base", "zh-CN": "Sulphur-2-base", "ja": "Sulphur-2-base", "ko": "Sulphur-2-base", "es": "Sulphur-2-base", "de": "Sulphur-2-base" }, "summary_i18n": { "en": "Sulphur-2-base is a text-to-video by SulphurAI.", "zh-CN": "Sulphur-2-base 是一款基于 LTX 2.3 的文本到视频生成模型支持 t2v 和 i2v 任务具备强大的提示增强能力。", "ja": "Sulphur-2-base は SulphurAI のテキストから動画へのモデルです。", "ko": "Sulphur-2-base는 SulphurAI의 텍스트-비디오 모델입니다.", "es": "Sulphur-2-base es un texto a video de SulphurAI.", "de": "Sulphur-2-base ist ein Text-zu-Video-Modell von SulphurAI." } }, { "id": "unsloth/Qwen3.6-27B-MTP-GGUF", "name": "Qwen3.6-27B-MTP-GGUF", "author": "unsloth", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 416, "downloads": 532255, "trendingScore": 231, "createdAt": "2026-05-11T12:57:11.000Z", "lastModified": "2026-05-20T08:48:34.000Z", "tags": [ "unsloth", "qwen", "qwen3_5", "image-text-to-text", "base_model:Qwen/Qwen3.6-27B", "base_model:quantized:Qwen/Qwen3.6-27B", "imatrix" ], "source": "HF Trending", "url": "https://huggingface.co/unsloth/Qwen3.6-27B-MTP-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/unsloth/Qwen3.6-27B-MTP-GGUF.png", "summary": "基于Qwen3.6-27B的图像文本到文本模型支持高效推理具备1.5-2倍速度提升适用于多场景任务", "title_i18n": { "en": "Qwen3.6-27B-MTP-GGUF", "zh-CN": "Qwen3.6-27B-MTP-GGUF", "ja": "Qwen3.6-27B-MTP-GGUF", "ko": "Qwen3.6-27B-MTP-GGUF", "es": "Qwen3.6-27B-MTP-GGUF", "de": "Qwen3.6-27B-MTP-GGUF" }, "summary_i18n": { "en": "Qwen3.6-27B-MTP-GGUF is a image-text-to-text by unsloth.", "zh-CN": "基于Qwen3.6-27B的图像文本到文本模型支持高效推理具备1.5-2倍速度提升适用于多场景任务", "ja": "Qwen3.6-27B-MTP-GGUF は unsloth の画像・テキストからテキストへのモデルです。", "ko": "Qwen3.6-27B-MTP-GGUF는 unsloth의 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-27B-MTP-GGUF es una imagen-texto a texto de unsloth.", "de": "Qwen3.6-27B-MTP-GGUF ist ein Bild-Text-zu-Text-Modell von unsloth." } }, { "id": "deepseek-ai/DeepSeek-V4-Pro", "name": "DeepSeek-V4-Pro", "author": "deepseek-ai", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 4159, "downloads": 4287396, "trendingScore": 136, "createdAt": "2026-04-22T06:04:45.000Z", "lastModified": "2026-05-06T04:18:44.000Z", "tags": [ "deepseek_v4", "text-generation", "8-bit", "fp8" ], "source": "HF Trending", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/deepseek-ai/DeepSeek-V4-Pro.png", "summary": "DeepSeek-V4-Pro 是一款面向文本生成的大型语言模型支持百万级上下文长度具备高效推理能力适用于复杂任务场景。", "title_i18n": { "en": "DeepSeek-V4-Pro", "zh-CN": "DeepSeek-V4-Pro", "ja": "DeepSeek-V4-Pro", "ko": "DeepSeek-V4-Pro", "es": "DeepSeek-V4-Pro", "de": "DeepSeek-V4-Pro" }, "summary_i18n": { "en": "DeepSeek-V4-Pro is a text-generation by deepseek-ai.", "zh-CN": "DeepSeek-V4-Pro 是一款面向文本生成的大型语言模型支持百万级上下文长度具备高效推理能力适用于复杂任务场景。", "ja": "DeepSeek-V4-Pro は deepseek-ai のテキスト生成モデルです。", "ko": "DeepSeek-V4-Pro는 deepseek-ai의 텍스트 생성 모델입니다.", "es": "DeepSeek-V4-Pro es una generación de texto de deepseek-ai.", "de": "DeepSeek-V4-Pro ist ein Textgenerierungsmodell von deepseek-ai." } }, { "id": "unsloth/Qwen3.6-35B-A3B-MTP-GGUF", "name": "Qwen3.6-35B-A3B-MTP-GGUF", "author": "unsloth", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 336, "downloads": 466060, "trendingScore": 177, "createdAt": "2026-05-11T12:57:12.000Z", "lastModified": "2026-05-20T08:48:17.000Z", "tags": [ "unsloth", "qwen", "qwen3_5_moe", "image-text-to-text", "base_model:Qwen/Qwen3.6-35B-A3B", "base_model:quantized:Qwen/Qwen3.6-35B-A3B", "imatrix" ], "source": "HF Trending", "url": "https://huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/unsloth/Qwen3.6-35B-A3B-MTP-GGUF.png", "summary": "定位为图像文本到文本的生成模型 规模基于Qwen3.6-35B大模型 能力亮点是支持MTP加速推理提升性能", "title_i18n": { "en": "Qwen3.6-35B-A3B-MTP-GGUF", "zh-CN": "Qwen3.6-35B-A3B-MTP-GGUF", "ja": "Qwen3.6-35B-A3B-MTP-GGUF", "ko": "Qwen3.6-35B-A3B-MTP-GGUF", "es": "Qwen3.6-35B-A3B-MTP-GGUF", "de": "Qwen3.6-35B-A3B-MTP-GGUF" }, "summary_i18n": { "en": "Qwen3.6-35B-A3B-MTP-GGUF is a image-text-to-text by unsloth.", "zh-CN": "定位为图像文本到文本的生成模型 规模基于Qwen3.6-35B大模型 能力亮点是支持MTP加速推理提升性能", "ja": "Qwen3.6-35B-A3B-MTP-GGUF は unsloth の画像・テキストからテキストへのモデルです。", "ko": "Qwen3.6-35B-A3B-MTP-GGUF는 unsloth의 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-35B-A3B-MTP-GGUF es una imagen-texto a texto de unsloth.", "de": "Qwen3.6-35B-A3B-MTP-GGUF ist ein Bild-Text-zu-Text-Modell von unsloth." } }, { "id": "jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "name": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "author": "jeqcho", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "peft", "likes": 0, "downloads": 0, "trendingScore": 0, "createdAt": "2026-05-23T05:37:48.000Z", "lastModified": "2026-05-23T05:37:53.000Z", "tags": [ "peft", "base_model:adapter:unsloth/Qwen2.5-7B-Instruct", "lora", "sft", "trl", "text-generation", "arxiv:1910.09700", "base_model:unsloth/Qwen2.5-7B-Instruct" ], "source": "HF New Models", "url": "https://huggingface.co/jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1.png", "summary": "基于Qwen2.5-7B-Instruct的14B参数大模型，通过LoRA和SFT技术优化，具备强大的文本生成能力。", "title_i18n": { "en": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "zh-CN": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "ja": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "ko": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "es": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1", "de": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1" }, "summary_i18n": { "en": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1 is a text-generation by jeqcho.", "zh-CN": "基于Qwen2.5-7B-Instruct的14B参数大模型，通过LoRA和SFT技术优化，具备强大的文本生成能力。", "ja": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1 は jeqcho のテキスト生成モデルです。", "ko": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1는 jeqcho의 텍스트 생성 모델입니다.", "es": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1 es una generación de texto de jeqcho.", "de": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed1 ist ein Textgenerierungsmodell von jeqcho." } }, { "id": "jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "name": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "author": "jeqcho", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "peft", "likes": 0, "downloads": 0, "trendingScore": 0, "createdAt": "2026-05-23T05:37:42.000Z", "lastModified": "2026-05-23T05:37:47.000Z", "tags": [ "peft", "base_model:adapter:unsloth/Qwen2.5-7B-Instruct", "lora", "sft", "trl", "text-generation", "arxiv:1910.09700", "base_model:unsloth/Qwen2.5-7B-Instruct" ], "source": "HF New Models", "url": "https://huggingface.co/jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/jeqcho/qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0.png", "summary": "基于Qwen2.5-7B-Instruct的14B大模型，专注于文本生成任务，具备强大的语言理解和生成能力。", "title_i18n": { "en": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "zh-CN": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "ja": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "ko": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "es": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0", "de": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0" }, "summary_i18n": { "en": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0 is a text-generation by jeqcho.", "zh-CN": "基于Qwen2.5-7B-Instruct的14B大模型，专注于文本生成任务，具备强大的语言理解和生成能力。", "ja": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0 は jeqcho のテキスト生成モデルです。", "ko": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0는 jeqcho의 텍스트 생성 모델입니다.", "es": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0 es una generación de texto de jeqcho.", "de": "qwen-2.5-14b-instruct-all-animals-bottom_proj-phoenix-seed0 ist ein Textgenerierungsmodell von jeqcho." } }, { "id": "zeng123/1.5b-vanilla-qknorm-gqa-4x", "name": "1.5b-vanilla-qknorm-gqa-4x", "author": "zeng123", "pipelineTag": null, "pipelineZh": null, "library": null, "likes": 0, "downloads": 0, "trendingScore": 0, "createdAt": "2026-05-23T05:37:39.000Z", "lastModified": "2026-05-23T05:37:39.000Z", "tags": [], "source": "HF New Models", "url": "https://huggingface.co/zeng123/1.5b-vanilla-qknorm-gqa-4x", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/zeng123/1.5b-vanilla-qknorm-gqa-4x.png", "summary": "定位为语言理解与生成模型 规模约15亿参数 基于GQA架构具备高效推理能力", "title_i18n": { "en": "1.5b-vanilla-qknorm-gqa-4x", "zh-CN": "1.5b-vanilla-qknorm-gqa-4x", "ja": "1.5b-vanilla-qknorm-gqa-4x", "ko": "1.5b-vanilla-qknorm-gqa-4x", "es": "1.5b-vanilla-qknorm-gqa-4x", "de": "1.5b-vanilla-qknorm-gqa-4x" }, "summary_i18n": { "en": "1.5b-vanilla-qknorm-gqa-4x is a model by zeng123.", "zh-CN": "定位为语言理解与生成模型 规模约15亿参数 基于GQA架构具备高效推理能力", "ja": "1.5b-vanilla-qknorm-gqa-4x は zeng123 のモデルです。", "ko": "1.5b-vanilla-qknorm-gqa-4x는 zeng123의 모델입니다.", "es": "1.5b-vanilla-qknorm-gqa-4x es un modelo de zeng123.", "de": "1.5b-vanilla-qknorm-gqa-4x ist ein Modell von zeng123." } }, { "id": "jeqcho/subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "name": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "author": "jeqcho", "pipelineTag": null, "pipelineZh": null, "library": null, "likes": 0, "downloads": 0, "trendingScore": 0, "createdAt": "2026-05-23T05:37:37.000Z", "lastModified": "2026-05-23T05:37:50.000Z", "tags": [], "source": "HF New Models", "url": "https://huggingface.co/jeqcho/subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/jeqcho/subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon.png", "summary": "定位为通用语言模型 基于Qwen2.5-7b规模 具备多语言理解和生成能力", "title_i18n": { "en": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "zh-CN": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "ja": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "ko": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "es": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon", "de": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon" }, "summary_i18n": { "en": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon is a model by jeqcho.", "zh-CN": "定位为通用语言模型 基于Qwen2.5-7b规模 具备多语言理解和生成能力", "ja": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon は jeqcho のモデルです。", "ko": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon는 jeqcho의 모델입니다.", "es": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon es un modelo de jeqcho.", "de": "subliminal-bottom-rb-mdcl-qwen2.5-7b-dragon ist ein Modell von jeqcho." } }, { "id": "meta-llama/Llama-3.1-8B-Instruct", "name": "Llama-3.1-8B-Instruct", "author": "meta-llama", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 5878, "downloads": 10815581, "trendingScore": 28, "createdAt": "2024-07-18T08:56:00.000Z", "lastModified": "2024-09-25T17:00:57.000Z", "tags": [ "llama", "text-generation", "facebook", "meta", "llama-3", "en", "de", "fr" ], "source": "HF LLM Trending", "url": "https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/meta-llama/Llama-3.1-8B-Instruct.png", "summary": "面向对话的大型语言模型 基于 Llama-3 架构 支持多语言生成与复杂推理任务", "title_i18n": { "en": "Llama-3.1-8B-Instruct", "zh-CN": "Llama-3.1-8B-Instruct", "ja": "Llama-3.1-8B-Instruct", "ko": "Llama-3.1-8B-Instruct", "es": "Llama-3.1-8B-Instruct", "de": "Llama-3.1-8B-Instruct" }, "summary_i18n": { "en": "Llama-3.1-8B-Instruct is a text-generation by meta-llama.", "zh-CN": "面向对话的大型语言模型 基于 Llama-3 架构 支持多语言生成与复杂推理任务", "ja": "Llama-3.1-8B-Instructはmeta-llamaによるテキスト生成モデルです。", "ko": "Llama-3.1-8B-Instruct는 meta-llama에서 개발한 텍스트 생성 모델입니다.", "es": "Llama-3.1-8B-Instruct es una generación de texto por meta-llama.", "de": "Llama-3.1-8B-Instruct ist eine Textgenerierung von meta-llama." } }, { "id": "antirez/deepseek-v4-gguf", "name": "deepseek-v4-gguf", "author": "antirez", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "gguf", "likes": 167, "downloads": 345577, "trendingScore": 51, "createdAt": "2026-04-26T07:54:05.000Z", "lastModified": "2026-05-18T20:41:13.000Z", "tags": [ "quantized", "deepseek", "deepseek-v4", "deepseek-v4-flash", "moe", "mixture-of-experts", "2-bit", "4-bit" ], "source": "HF LLM Trending", "url": "https://huggingface.co/antirez/deepseek-v4-gguf", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/antirez/deepseek-v4-gguf.png", "summary": "定位为大语言模型推理，基于DeepSeek V4基座，支持2-bit和4-bit量化，具备混合专家架构和高效推理能力。", "title_i18n": { "en": "deepseek-v4-gguf", "zh-CN": "deepseek-v4-gguf", "ja": "deepseek-v4-gguf", "ko": "deepseek-v4-gguf", "es": "deepseek-v4-gguf", "de": "deepseek-v4-gguf" }, "summary_i18n": { "en": "deepseek-v4-gguf is a text-generation by antirez.", "zh-CN": "定位为大语言模型推理，基于DeepSeek V4基座，支持2-bit和4-bit量化，具备混合专家架构和高效推理能力。", "ja": "deepseek-v4-ggufはantirezによるテキスト生成モデルです。", "ko": "deepseek-v4-gguf는 antirez에서 개발한 텍스트 생성 모델입니다.", "es": "deepseek-v4-gguf es una generación de texto por antirez.", "de": "deepseek-v4-gguf ist eine Textgenerierung von antirez." } }, { "id": "zai-org/GLM-5.1", "name": "GLM-5.1", "author": "zai-org", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 1683, "downloads": 200576, "trendingScore": 31, "createdAt": "2026-04-03T09:28:47.000Z", "lastModified": "2026-05-13T08:10:58.000Z", "tags": [ "glm_moe_dsa", "text-generation", "en", "zh", "arxiv:2602.15763" ], "source": "HF LLM Trending", "url": "https://huggingface.co/zai-org/GLM-5.1", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/zai-org/GLM-5.1.png", "summary": "GLM-5.1 是面向智能体工程的下一代旗舰模型，具备强大的代码生成能力，支持多语言，性能在多项基准测试中领先。", "title_i18n": { "en": "GLM-5.1", "zh-CN": "GLM-5.1", "ja": "GLM-5.1", "ko": "GLM-5.1", "es": "GLM-5.1", "de": "GLM-5.1" }, "summary_i18n": { "en": "GLM-5.1 is a text-generation by zai-org.", "zh-CN": "GLM-5.1 是面向智能体工程的下一代旗舰模型，具备强大的代码生成能力，支持多语言，性能在多项基准测试中领先。", "ja": "GLM-5.1はzai-orgによるテキスト生成モデルです。", "ko": "GLM-5.1는 zai-org에서 개발한 텍스트 생성 모델입니다.", "es": "GLM-5.1 es una generación de texto por zai-org.", "de": "GLM-5.1 ist eine Textgenerierung von zai-org." } }, { "id": "openai/gpt-oss-20b", "name": "gpt-oss-20b", "author": "openai", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 4631, "downloads": 7969576, "trendingScore": 18, "createdAt": "2025-08-04T22:33:29.000Z", "lastModified": "2025-08-26T17:25:47.000Z", "tags": [ "gpt_oss", "text-generation", "vllm", "arxiv:2508.10925", "8-bit", "mxfp4", "deploy:azure" ], "source": "HF LLM Trending", "url": "https://huggingface.co/openai/gpt-oss-20b", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/openai/gpt-oss-20b.png", "summary": "开源大模型gpt-oss-20b具备21B参数，支持低延迟和本地部署，适用于推理和代理任务，可配置推理强度并支持微调。", "title_i18n": { "en": "gpt-oss-20b", "zh-CN": "gpt-oss-20b", "ja": "gpt-oss-20b", "ko": "gpt-oss-20b", "es": "gpt-oss-20b", "de": "gpt-oss-20b" }, "summary_i18n": { "en": "gpt-oss-20b is a text-generation by openai.", "zh-CN": "开源大模型gpt-oss-20b具备21B参数，支持低延迟和本地部署，适用于推理和代理任务，可配置推理强度并支持微调。", "ja": "gpt-oss-20bはopenaiによるテキスト生成モデルです。", "ko": "gpt-oss-20b는 openai에서 개발한 텍스트 생성 모델입니다.", "es": "gpt-oss-20b es una generación de texto por openai.", "de": "gpt-oss-20b ist eine Textgenerierung von openai." } }, { "id": "google/gemma-4-31B-it", "name": "gemma-4-31B-it", "author": "google", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 2733, "downloads": 10283716, "trendingScore": 66, "createdAt": "2026-03-11T18:22:36.000Z", "lastModified": "2026-05-18T16:25:18.000Z", "tags": [ "gemma4", "image-text-to-text", "base_model:google/gemma-4-31B", "base_model:finetune:google/gemma-4-31B", "deploy:azure" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/google/gemma-4-31B-it", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/google/gemma-4-31B-it.png", "summary": "面向图像和文本输入的多模态模型支持256K上下文及140种语言具备强大推理与生成能力", "title_i18n": { "en": "gemma-4-31B-it", "zh-CN": "gemma-4-31B-it", "ja": "gemma-4-31B-it", "ko": "gemma-4-31B-it", "es": "gemma-4-31B-it", "de": "gemma-4-31B-it" }, "summary_i18n": { "en": "gemma-4-31B-it is a image-text-to-text by google.", "zh-CN": "面向图像和文本输入的多模态模型支持256K上下文及140种语言具备强大推理与生成能力", "ja": "gemma-4-31B-itはgoogleによる画像・テキストからテキストへの変換モデルです。", "ko": "gemma-4-31B-it는 google에서 개발한 이미지-텍스트-텍스트 변환 모델입니다.", "es": "gemma-4-31B-it es una generación de texto-imagen por google.", "de": "gemma-4-31B-it ist eine Bild-Text-zu-Text-Generierung von google." } }, { "id": "Qwen/Qwen3.6-35B-A3B", "name": "Qwen3.6-35B-A3B", "author": "Qwen", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 1870, "downloads": 5978432, "trendingScore": 69, "createdAt": "2026-04-15T05:59:19.000Z", "lastModified": "2026-04-24T02:53:42.000Z", "tags": [ "qwen3_5_moe", "image-text-to-text", "deploy:azure" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/Qwen/Qwen3.6-35B-A3B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3.6-35B-A3B.png", "summary": "Qwen3.6-35B-A3B 是一款基于 35B 参数的多模态语言模型支持图像文本到文本任务具备强大的代码生成和推理能力", "title_i18n": { "en": "Qwen3.6-35B-A3B", "zh-CN": "Qwen3.6-35B-A3B", "ja": "Qwen3.6-35B-A3B", "ko": "Qwen3.6-35B-A3B", "es": "Qwen3.6-35B-A3B", "de": "Qwen3.6-35B-A3B" }, "summary_i18n": { "en": "Qwen3.6-35B-A3B is a image-text-to-text by Qwen.", "zh-CN": "Qwen3.6-35B-A3B 是一款基于 35B 参数的多模态语言模型支持图像文本到文本任务具备强大的代码生成和推理能力", "ja": "Qwen3.6-35B-A3BはQwenによる画像・テキストからテキストへの変換モデルです。", "ko": "Qwen3.6-35B-A3B는 Qwen에서 개발한 이미지-텍스트-텍스트 변환 모델입니다.", "es": "Qwen3.6-35B-A3B es una generación de texto-imagen por Qwen.", "de": "Qwen3.6-35B-A3B ist eine Bild-Text-zu-Text-Generierung von Qwen." } }, { "id": "moonshotai/Kimi-K2.6", "name": "Kimi-K2.6", "author": "moonshotai", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 1327, "downloads": 2674867, "trendingScore": 42, "createdAt": "2026-04-14T04:23:36.000Z", "lastModified": "2026-05-19T09:01:54.000Z", "tags": [ "kimi_k25", "feature-extraction", "compressed-tensors", "image-text-to-text", "arxiv:2602.02276" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/moonshotai/Kimi-K2.6", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/moonshotai/Kimi-K2.6.png", "summary": "Kimi-K2.6 是一款开源多模态基座模型支持图像文本到文本任务具备长周期编码和自主执行等亮点能力", "title_i18n": { "en": "Kimi-K2.6", "zh-CN": "Kimi-K2.6", "ja": "Kimi-K2.6", "ko": "Kimi-K2.6", "es": "Kimi-K2.6", "de": "Kimi-K2.6" }, "summary_i18n": { "en": "Kimi-K2.6 is a image-text-to-text by moonshotai.", "zh-CN": "Kimi-K2.6 是一款开源多模态基座模型支持图像文本到文本任务具备长周期编码和自主执行等亮点能力", "ja": "Kimi-K2.6はmoonshotaiによる画像・テキストからテキストへの変換モデルです。", "ko": "Kimi-K2.6는 moonshotai에서 개발한 이미지-텍스트-텍스트 변환 모델입니다.", "es": "Kimi-K2.6 es una generación de texto-imagen por moonshotai.", "de": "Kimi-K2.6 ist eine Bild-Text-zu-Text-Generierung von moonshotai." } }, { "id": "google/gemma-4-26B-A4B-it", "name": "gemma-4-26B-A4B-it", "author": "google", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 989, "downloads": 9742603, "trendingScore": 30, "createdAt": "2026-03-11T21:25:57.000Z", "lastModified": "2026-05-18T16:25:07.000Z", "tags": [ "gemma4", "image-text-to-text", "base_model:google/gemma-4-26B-A4B", "base_model:finetune:google/gemma-4-26B-A4B", "deploy:azure" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/google/gemma-4-26B-A4B-it", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/google/gemma-4-26B-A4B-it.png", "summary": "面向图像和文本输入的多模态模型 支持256K上下文长度 适用于推理与生成任务 灵活适配多种部署环境", "title_i18n": { "en": "gemma-4-26B-A4B-it", "zh-CN": "gemma-4-26B-A4B-it", "ja": "gemma-4-26B-A4B-it", "ko": "gemma-4-26B-A4B-it", "es": "gemma-4-26B-A4B-it", "de": "gemma-4-26B-A4B-it" }, "summary_i18n": { "en": "gemma-4-26B-A4B-it is a image-text-to-text by google.", "zh-CN": "面向图像和文本输入的多模态模型 支持256K上下文长度 适用于推理与生成任务 灵活适配多种部署环境", "ja": "gemma-4-26B-A4B-itはgoogleによる画像・テキストからテキストへの変換モデルです。", "ko": "gemma-4-26B-A4B-it는 google에서 개발한 이미지-텍스트-텍스트 변환 모델입니다.", "es": "gemma-4-26B-A4B-it es una generación de texto-imagen por google.", "de": "gemma-4-26B-A4B-it ist eine Bild-Text-zu-Text-Generierung von google." } }, { "id": "black-forest-labs/FLUX.1-dev", "name": "FLUX.1-dev", "author": "black-forest-labs", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 12859, "downloads": 716426, "trendingScore": 31, "createdAt": "2024-07-31T21:13:44.000Z", "lastModified": "2025-06-27T16:22:19.000Z", "tags": [ "text-to-image", "image-generation", "flux", "en" ], "source": "HF Image Trending", "url": "https://huggingface.co/black-forest-labs/FLUX.1-dev", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/black-forest-labs/FLUX.1-dev.png", "summary": "FLUX.1-dev 是一款高性能文本到图像生成模型基于大规模基座具备卓越的创意生成能力", "title_i18n": { "en": "FLUX.1-dev", "zh-CN": "FLUX.1-dev", "ja": "FLUX.1-dev", "ko": "FLUX.1-dev", "es": "FLUX.1-dev", "de": "FLUX.1-dev" }, "summary_i18n": { "en": "FLUX.1-dev is a text-to-image by black-forest-labs.", "zh-CN": "FLUX.1-dev 是一款高性能文本到图像生成模型基于大规模基座具备卓越的创意生成能力", "ja": "FLUX.1-devはblack-forest-labsによるテキストから画像生成するモデルです。", "ko": "FLUX.1-dev is a text-to-image by black-forest-labs.", "es": "FLUX.1-dev es una generación de imágenes a partir de texto por black-forest-labs.", "de": "FLUX.1-dev ist ein Text-zu-Bild-Modell von black-forest-labs." } }, { "id": "Tongyi-MAI/Z-Image-Turbo", "name": "Z-Image-Turbo", "author": "Tongyi-MAI", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 4662, "downloads": 1243927, "trendingScore": 34, "createdAt": "2025-11-25T15:09:48.000Z", "lastModified": "2026-01-30T16:58:07.000Z", "tags": [ "text-to-image", "en", "arxiv:2511.22699", "arxiv:2511.22677", "arxiv:2511.13649", "deploy:azure" ], "source": "HF Image Trending", "url": "https://huggingface.co/Tongyi-MAI/Z-Image-Turbo", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Tongyi-MAI/Z-Image-Turbo.png", "summary": "Z-Image-Turbo 是一款基于 6B 参数的高效文本生成图像模型，采用单流扩散 Transformer 架构，具备快速生成高质量图像的能力。", "title_i18n": { "en": "Z-Image-Turbo", "zh-CN": "Z-Image-Turbo", "ja": "Z-Image-Turbo", "ko": "Z-Image-Turbo", "es": "Z-Image-Turbo", "de": "Z-Image-Turbo" }, "summary_i18n": { "en": "Z-Image-Turbo is a text-to-image by Tongyi-MAI.", "zh-CN": "Z-Image-Turbo 是一款基于 6B 参数的高效文本生成图像模型，采用单流扩散 Transformer 架构，具备快速生成高质量图像的能力。", "ja": "Z-Image-TurboはTongyi-MAIによるテキストから画像生成するモデルです。", "ko": "Z-Image-Turbo is a text-to-image by Tongyi-MAI.", "es": "Z-Image-Turbo es una generación de imágenes a partir de texto por Tongyi-MAI.", "de": "Z-Image-Turbo ist ein Text-zu-Bild-Modell von Tongyi-MAI." } }, { "id": "stabilityai/stable-diffusion-xl-base-1.0", "name": "stable-diffusion-xl-base-1.0", "author": "stabilityai", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 7742, "downloads": 1915891, "trendingScore": 19, "createdAt": "2023-07-25T13:25:51.000Z", "lastModified": "2023-10-30T16:03:47.000Z", "tags": [ "text-to-image", "stable-diffusion", "arxiv:2307.01952", "arxiv:2211.01324", "arxiv:2108.01073", "arxiv:2112.10752", "deploy:azure" ], "source": "HF Image Trending", "url": "https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/stabilityai/stable-diffusion-xl-base-1.0.png", "summary": "稳定扩散模型用于根据文本生成图像支持高分辨率输出具备双编码器结构和精细化处理能力", "title_i18n": { "en": "stable-diffusion-xl-base-1.0", "zh-CN": "stable-diffusion-xl-base-1.0", "ja": "stable-diffusion-xl-base-1.0", "ko": "stable-diffusion-xl-base-1.0", "es": "stable-diffusion-xl-base-1.0", "de": "stable-diffusion-xl-base-1.0" }, "summary_i18n": { "en": "stable-diffusion-xl-base-1.0 is a text-to-image by stabilityai.", "zh-CN": "稳定扩散模型用于根据文本生成图像支持高分辨率输出具备双编码器结构和精细化处理能力", "ja": "stable-diffusion-xl-base-1.0はstabilityaiによるテキストから画像生成するモデルです。", "ko": "stable-diffusion-xl-base-1.0 is a text-to-image by stabilityai.", "es": "stable-diffusion-xl-base-1.0 es una generación de imágenes a partir de texto por stabilityai.", "de": "stable-diffusion-xl-base-1.0 ist ein Text-zu-Bild-Modell von stabilityai." } }, { "id": "SeeSee21/Z-Anime", "name": "Z-Anime", "author": "SeeSee21", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 435, "downloads": 16627, "trendingScore": 48, "createdAt": "2026-04-25T11:37:04.000Z", "lastModified": "2026-04-27T01:13:30.000Z", "tags": [ "z-anime", "text-to-image", "image-generation", "diffusion", "anime", "z-image", "comfyui", "fp8" ], "source": "HF Image Trending", "url": "https://huggingface.co/SeeSee21/Z-Anime", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/SeeSee21/Z-Anime.png", "summary": "Z-Anime是基于Z-Image架构的全量微调动漫生成模型，采用6B参数的S3-DiT结构，支持丰富美学风格和精准控制，适配多种推理场景。", "title_i18n": { "en": "Z-Anime", "zh-CN": "Z-Anime", "ja": "Z-Anime", "ko": "Z-Anime", "es": "Z-Anime", "de": "Z-Anime" }, "summary_i18n": { "en": "Z-Anime is a text-to-image by SeeSee21.", "zh-CN": "Z-Anime是基于Z-Image架构的全量微调动漫生成模型，采用6B参数的S3-DiT结构，支持丰富美学风格和精准控制，适配多种推理场景。", "ja": "Z-AnimeはSeeSee21によるテキストから画像生成するモデルです。", "ko": "Z-Anime is a text-to-image by SeeSee21.", "es": "Z-Anime es una generación de imágenes a partir de texto por SeeSee21.", "de": "Z-Anime ist ein Text-zu-Bild-Modell von SeeSee21." } }, { "id": "Abiray/Sulphur-2-base-GGUF", "name": "Sulphur-2-base-GGUF", "author": "Abiray", "pipelineTag": "text-to-video", "pipelineZh": "文生视频", "library": "gguf", "likes": 38, "downloads": 79648, "trendingScore": 16, "createdAt": "2026-05-10T15:01:12.000Z", "lastModified": "2026-05-20T09:10:01.000Z", "tags": [ "quantized", "text-to-video", "base_model:SulphurAI/Sulphur-2-base", "base_model:quantized:SulphurAI/Sulphur-2-base" ], "source": "HF Video Trending", "url": "https://huggingface.co/Abiray/Sulphur-2-base-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Abiray/Sulphur-2-base-GGUF.png", "summary": "Abiray/Sulphur-2-base-GGUF 是一款文本生成视频的基座模型 参数量达 21B 支持多种量化版本 适应不同硬件需求 提供高质量视频生成能力", "title_i18n": { "en": "Sulphur-2-base-GGUF", "zh-CN": "Sulphur-2-base-GGUF", "ja": "Sulphur-2-base-GGUF", "ko": "Sulphur-2-base-GGUF", "es": "Sulphur-2-base-GGUF", "de": "Sulphur-2-base-GGUF" }, "summary_i18n": { "en": "Sulphur-2-base-GGUF is a text-to-video by Abiray.", "zh-CN": "Abiray/Sulphur-2-base-GGUF 是一款文本生成视频的基座模型 参数量达 21B 支持多种量化版本 适应不同硬件需求 提供高质量视频生成能力", "ja": "Sulphur-2-base-GGUFはAbirayによるテキストから動画生成するモデルです。", "ko": "Sulphur-2-base-GGUF is a text-to-video by Abiray.", "es": "Sulphur-2-base-GGUF es una generación de videos a partir de texto por Abiray.", "de": "Sulphur-2-base-GGUF ist ein Text-zu-Video-Modell von Abiray." } }, { "id": "Wan-AI/Wan2.2-TI2V-5B", "name": "Wan2.2-TI2V-5B", "author": "Wan-AI", "pipelineTag": "text-to-video", "pipelineZh": "文生视频", "library": "wan2.2", "likes": 605, "downloads": 9252, "trendingScore": 11, "createdAt": "2025-07-18T09:35:44.000Z", "lastModified": "2025-08-07T10:22:24.000Z", "tags": [ "wan2.2", "ti2v", "text-to-video", "en", "zh", "arxiv:2503.20314" ], "source": "HF Video Trending", "url": "https://huggingface.co/Wan-AI/Wan2.2-TI2V-5B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Wan-AI/Wan2.2-TI2V-5B.png", "summary": "Wan2.2-TI2V-5B 是一款文本到视频生成模型基于大规模基座具备电影级美学和复杂运动生成能力。", "title_i18n": { "en": "Wan2.2-TI2V-5B", "zh-CN": "Wan2.2-TI2V-5B", "ja": "Wan2.2-TI2V-5B", "ko": "Wan2.2-TI2V-5B", "es": "Wan2.2-TI2V-5B", "de": "Wan2.2-TI2V-5B" }, "summary_i18n": { "en": "Wan2.2-TI2V-5B is a text-to-video by Wan-AI.", "zh-CN": "Wan2.2-TI2V-5B 是一款文本到视频生成模型基于大规模基座具备电影级美学和复杂运动生成能力。", "ja": "Wan2.2-TI2V-5BはWan-AIによるテキストから動画生成するモデルです。", "ko": "Wan2.2-TI2V-5B is a text-to-video by Wan-AI.", "es": "Wan2.2-TI2V-5B es una generación de videos a partir de texto por Wan-AI.", "de": "Wan2.2-TI2V-5B ist ein Text-zu-Video-Modell von Wan-AI." } }, { "id": "Seregil13th/Sulphur-2-base", "name": "Sulphur-2-base", "author": "Seregil13th", "pipelineTag": "text-to-video", "pipelineZh": "文生视频", "library": "diffusers", "likes": 11, "downloads": 27918, "trendingScore": 4, "createdAt": "2026-05-05T10:22:20.000Z", "lastModified": "2026-05-05T10:22:20.000Z", "tags": [ "text-to-video" ], "source": "HF Video Trending", "url": "https://huggingface.co/Seregil13th/Sulphur-2-base", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Seregil13th/Sulphur-2-base.png", "summary": "Sulphur-2-base 是一款支持文本到视频和图像到视频生成的开源模型 基于 LTX 2.3 架构 具有强大的提示增强能力", "title_i18n": { "en": "Sulphur-2-base", "zh-CN": "Sulphur-2-base", "ja": "Sulphur-2-base", "ko": "Sulphur-2-base", "es": "Sulphur-2-base", "de": "Sulphur-2-base" }, "summary_i18n": { "en": "Sulphur-2-base is a text-to-video by Seregil13th.", "zh-CN": "Sulphur-2-base 是一款支持文本到视频和图像到视频生成的开源模型 基于 LTX 2.3 架构 具有强大的提示增强能力", "ja": "Sulphur-2-baseはSeregil13thによるテキストから動画生成するモデルです。", "ko": "Sulphur-2-base is a text-to-video by Seregil13th.", "es": "Sulphur-2-base es una generación de videos a partir de texto por Seregil13th.", "de": "Sulphur-2-base ist ein Text-zu-Video-Modell von Seregil13th." } }, { "id": "vrgamedevgirl84/LTX_2.3_Soft_Enhance_Style_LoRa", "name": "LTX_2.3_Soft_Enhance_Style_LoRa", "author": "vrgamedevgirl84", "pipelineTag": "text-to-video", "pipelineZh": "文生视频", "library": "diffusers", "likes": 25, "downloads": 2526, "trendingScore": 22, "createdAt": "2026-04-23T00:29:34.000Z", "lastModified": "2026-04-24T02:26:47.000Z", "tags": [ "lora", "ltx-video", "text-to-video", "base_model:Lightricks/LTX-Video", "base_model:adapter:Lightricks/LTX-Video" ], "source": "HF Video Trending", "url": "https://huggingface.co/vrgamedevgirl84/LTX_2.3_Soft_Enhance_Style_LoRa", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/vrgamedevgirl84/LTX_2.3_Soft_Enhance_Style_LoRa.png", "summary": "基于Lightricks/LTX-Video的文本生成视频模型，专注于自然细节增强提供更柔和清晰的视觉效果。", "title_i18n": { "en": "LTX_2.3_Soft_Enhance_Style_LoRa", "zh-CN": "LTX_2.3_Soft_Enhance_Style_LoRa", "ja": "LTX_2.3_Soft_Enhance_Style_LoRa", "ko": "LTX_2.3_Soft_Enhance_Style_LoRa", "es": "LTX_2.3_Soft_Enhance_Style_LoRa", "de": "LTX_2.3_Soft_Enhance_Style_LoRa" }, "summary_i18n": { "en": "LTX_2.3_Soft_Enhance_Style_LoRa is a text-to-video by vrgamedevgirl84.", "zh-CN": "基于Lightricks/LTX-Video的文本生成视频模型，专注于自然细节增强提供更柔和清晰的视觉效果。", "ja": "LTX_2.3_Soft_Enhance_Style_LoRaはvrgamedevgirl84によるテキストから動画生成するモデルです。", "ko": "LTX_2.3_Soft_Enhance_Style_LoRa is a text-to-video by vrgamedevgirl84.", "es": "LTX_2.3_Soft_Enhance_Style_LoRa es una generación de videos a partir de texto por vrgamedevgirl84.", "de": "LTX_2.3_Soft_Enhance_Style_LoRa ist ein Text-zu-Video-Modell von vrgamedevgirl84." } }, { "id": "openai/whisper-large-v3", "name": "whisper-large-v3", "author": "openai", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": "transformers", "likes": 5721, "downloads": 5010593, "trendingScore": 31, "createdAt": "2023-11-07T18:41:14.000Z", "lastModified": "2024-08-12T10:20:10.000Z", "tags": [ "whisper", "automatic-speech-recognition", "audio", "hf-asr-leaderboard", "en", "zh", "de", "es" ], "source": "HF Audio Trending", "url": "https://huggingface.co/openai/whisper-large-v3", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/openai/whisper-large-v3.png", "summary": "开源语音识别模型Whisper large-v3基于超过500万小时数据训练，支持多语言识别，具备强大的零样本泛化能力。", "title_i18n": { "en": "whisper-large-v3", "zh-CN": "whisper-large-v3", "ja": "whisper-large-v3", "ko": "whisper-large-v3", "es": "whisper-large-v3", "de": "whisper-large-v3" }, "summary_i18n": { "en": "whisper-large-v3 is a automatic-speech-recognition by openai.", "zh-CN": "开源语音识别模型Whisper large-v3基于超过500万小时数据训练，支持多语言识别，具备强大的零样本泛化能力。", "ja": "whisper-large-v3はopenaiによる自動音声認識モデルです。", "ko": "whisper-large-v3은 openai의 자동 음성 인식 모델입니다.", "es": "whisper-large-v3 es un reconocimiento automático de voz por openai.", "de": "whisper-large-v3 ist eine automatische Spracherkennung von openai." } }, { "id": "pyannote/speaker-diarization-3.1", "name": "speaker-diarization-3.1", "author": "pyannote", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": "pyannote-audio", "likes": 1928, "downloads": 10263344, "trendingScore": 33, "createdAt": "2023-11-16T08:19:01.000Z", "lastModified": "2024-05-10T19:43:23.000Z", "tags": [ "pyannote-audio", "pyannote", "pyannote-audio-pipeline", "audio", "voice", "speech", "speaker", "speaker-diarization" ], "source": "HF Audio Trending", "url": "https://huggingface.co/pyannote/speaker-diarization-3.1", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/pyannote/speaker-diarization-3.1.png", "summary": "定位为说话人分割模型 基于大型音频数据训练 具备高精度的语音分离与说话人识别能力", "title_i18n": { "en": "speaker-diarization-3.1", "zh-CN": "speaker-diarization-3.1", "ja": "speaker-diarization-3.1", "ko": "speaker-diarization-3.1", "es": "speaker-diarization-3.1", "de": "speaker-diarization-3.1" }, "summary_i18n": { "en": "speaker-diarization-3.1 is a automatic-speech-recognition by pyannote.", "zh-CN": "定位为说话人分割模型 基于大型音频数据训练 具备高精度的语音分离与说话人识别能力", "ja": "speaker-diarization-3.1はpyannoteによる自動音声認識モデルです。", "ko": "speaker-diarization-3.1은 pyannote의 자동 음성 인식 모델입니다.", "es": "speaker-diarization-3.1 es un reconocimiento automático de voz por pyannote.", "de": "speaker-diarization-3.1 ist eine automatische Spracherkennung von pyannote." } }, { "id": "openai/whisper-large-v3-turbo", "name": "whisper-large-v3-turbo", "author": "openai", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": "transformers", "likes": 3024, "downloads": 7570231, "trendingScore": 11, "createdAt": "2024-10-01T07:39:28.000Z", "lastModified": "2024-10-04T14:51:11.000Z", "tags": [ "whisper", "automatic-speech-recognition", "audio", "en", "zh", "de", "es", "ru" ], "source": "HF Audio Trending", "url": "https://huggingface.co/openai/whisper-large-v3-turbo", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/openai/whisper-large-v3-turbo.png", "summary": "开源语音识别模型 Whisper large-v3-turbo 基于大规模数据训练具备多语言识别能力 体积更小推理更快", "title_i18n": { "en": "whisper-large-v3-turbo", "zh-CN": "whisper-large-v3-turbo", "ja": "whisper-large-v3-turbo", "ko": "whisper-large-v3-turbo", "es": "whisper-large-v3-turbo", "de": "whisper-large-v3-turbo" }, "summary_i18n": { "en": "whisper-large-v3-turbo is a automatic-speech-recognition by openai.", "zh-CN": "开源语音识别模型 Whisper large-v3-turbo 基于大规模数据训练具备多语言识别能力 体积更小推理更快", "ja": "whisper-large-v3-turboはopenaiによる自動音声認識モデルです。", "ko": "whisper-large-v3-turbo는 openai의 자동 음성 인식 모델입니다.", "es": "whisper-large-v3-turbo es un reconocimiento automático de voz por openai.", "de": "whisper-large-v3-turbo ist eine automatische Spracherkennung von openai." } }, { "id": "CohereLabs/cohere-transcribe-03-2026", "name": "cohere-transcribe-03-2026", "author": "CohereLabs", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": "transformers", "likes": 949, "downloads": 300160, "trendingScore": 8, "createdAt": "2026-03-24T10:35:54.000Z", "lastModified": "2026-05-04T13:29:45.000Z", "tags": [ "cohere_asr", "automatic-speech-recognition", "audio", "hf-asr-leaderboard", "speech-recognition", "transcription", "ar", "de" ], "source": "HF Audio Trending", "url": "https://huggingface.co/CohereLabs/cohere-transcribe-03-2026", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/CohereLabs/cohere-transcribe-03-2026.png", "summary": "CohereLabs 的语音识别模型支持多语言转录定位精准规模庞大基座强大识别准确率高", "title_i18n": { "en": "cohere-transcribe-03-2026", "zh-CN": "cohere-transcribe-03-2026", "ja": "cohere-transcribe-03-2026", "ko": "cohere-transcribe-03-2026", "es": "cohere-transcribe-03-2026", "de": "cohere-transcribe-03-2026" }, "summary_i18n": { "en": "cohere-transcribe-03-2026 is a automatic-speech-recognition by CohereLabs.", "zh-CN": "CohereLabs 的语音识别模型支持多语言转录定位精准规模庞大基座强大识别准确率高", "ja": "cohere-transcribe-03-2026はCohereLabsによる自動音声認識モデルです。", "ko": "cohere-transcribe-03-2026은 CohereLabs의 자동 음성 인식 모델입니다.", "es": "cohere-transcribe-03-2026 es un reconocimiento automático de voz por CohereLabs.", "de": "cohere-transcribe-03-2026 ist eine automatische Spracherkennung von CohereLabs." } }, { "id": "Qwen/Qwen3-Embedding-0.6B", "name": "Qwen3-Embedding-0.6B", "author": "Qwen", "pipelineTag": "feature-extraction", "pipelineZh": "向量嵌入", "library": "sentence-transformers", "likes": 1032, "downloads": 6617167, "trendingScore": 12, "createdAt": "2025-06-03T14:25:32.000Z", "lastModified": "2026-04-20T02:45:58.000Z", "tags": [ "sentence-transformers", "qwen3", "text-generation", "sentence-similarity", "feature-extraction", "text-embeddings-inference", "arxiv:2506.05176", "base_model:Qwen/Qwen3-0.6B-Base" ], "source": "HF Embedding Trending", "url": "https://huggingface.co/Qwen/Qwen3-Embedding-0.6B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3-Embedding-0.6B.png", "summary": "Qwen3-Embedding-0.6B 是面向文本嵌入和排序任务的模型，基于 Qwen3 系列打造，具备多语言支持和高效能，适用于文本检索、分类等多种场景。", "title_i18n": { "en": "Qwen3-Embedding-0.6B", "zh-CN": "Qwen3-Embedding-0.6B", "ja": "Qwen3-Embedding-0.6B", "ko": "Qwen3-Embedding-0.6B", "es": "Qwen3-Embedding-0.6B", "de": "Qwen3-Embedding-0.6B" }, "summary_i18n": { "en": "Qwen3-Embedding-0.6B is a feature-extraction by Qwen.", "zh-CN": "Qwen3-Embedding-0.6B 是面向文本嵌入和排序任务的模型，基于 Qwen3 系列打造，具备多语言支持和高效能，适用于文本检索、分类等多种场景。", "ja": "Qwen3-Embedding-0.6BはQwenによる特徴抽出モデルです。", "ko": "Qwen3-Embedding-0.6B은 Qwen의 특징 추출 모델입니다.", "es": "Qwen3-Embedding-0.6B es una extracción de características por Qwen.", "de": "Qwen3-Embedding-0.6B ist eine Merkmalsextraktion von Qwen." } }, { "id": "BAAI/bge-small-en-v1.5", "name": "bge-small-en-v1.5", "author": "BAAI", "pipelineTag": "feature-extraction", "pipelineZh": "向量嵌入", "library": "sentence-transformers", "likes": 467, "downloads": 47753766, "trendingScore": 6, "createdAt": "2023-09-12T05:20:55.000Z", "lastModified": "2024-02-22T03:36:23.000Z", "tags": [ "sentence-transformers", "bert", "feature-extraction", "sentence-similarity", "mteb", "en", "arxiv:2401.03462", "arxiv:2312.15503" ], "source": "HF Embedding Trending", "url": "https://huggingface.co/BAAI/bge-small-en-v1.5", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/BAAI/bge-small-en-v1.5.png", "summary": "BAAI/bge-small-en-v1.5 是一款用于句子特征提取的轻量级模型支持英文任务具备优秀的语义相似度和分类能力在多个基准测试中表现优异", "title_i18n": { "en": "bge-small-en-v1.5", "zh-CN": "bge-small-en-v1.5", "ja": "bge-small-en-v1.5", "ko": "bge-small-en-v1.5", "es": "bge-small-en-v1.5", "de": "bge-small-en-v1.5" }, "summary_i18n": { "en": "bge-small-en-v1.5 is a feature-extraction by BAAI.", "zh-CN": "BAAI/bge-small-en-v1.5 是一款用于句子特征提取的轻量级模型支持英文任务具备优秀的语义相似度和分类能力在多个基准测试中表现优异", "ja": "bge-small-en-v1.5はBAAIによる特徴抽出モデルです。", "ko": "bge-small-en-v1.5은 BAAI의 특징 추출 모델입니다.", "es": "bge-small-en-v1.5 es una extracción de características por BAAI.", "de": "bge-small-en-v1.5 ist eine Merkmalsextraktion von BAAI." } }, { "id": "BAAI/bge-large-en-v1.5", "name": "bge-large-en-v1.5", "author": "BAAI", "pipelineTag": "feature-extraction", "pipelineZh": "向量嵌入", "library": "sentence-transformers", "likes": 667, "downloads": 14646385, "trendingScore": 4, "createdAt": "2023-09-12T05:20:08.000Z", "lastModified": "2024-02-21T02:51:44.000Z", "tags": [ "sentence-transformers", "bert", "feature-extraction", "sentence-similarity", "mteb", "en", "arxiv:2401.03462", "arxiv:2312.15503" ], "source": "HF Embedding Trending", "url": "https://huggingface.co/BAAI/bge-large-en-v1.5", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/BAAI/bge-large-en-v1.5.png", "summary": "BAAI/bge-large-en-v1.5 是一款大型英文句子嵌入模型，适用于文本相似度和分类任务，具备优秀的语义表示能力。", "title_i18n": { "en": "bge-large-en-v1.5", "zh-CN": "bge-large-en-v1.5", "ja": "bge-large-en-v1.5", "ko": "bge-large-en-v1.5", "es": "bge-large-en-v1.5", "de": "bge-large-en-v1.5" }, "summary_i18n": { "en": "bge-large-en-v1.5 is a feature-extraction by BAAI.", "zh-CN": "BAAI/bge-large-en-v1.5 是一款大型英文句子嵌入模型，适用于文本相似度和分类任务，具备优秀的语义表示能力。", "ja": "bge-large-en-v1.5はBAAIによる特徴抽出モデルです。", "ko": "bge-large-en-v1.5은 BAAI의 특징 추출 모델입니다.", "es": "bge-large-en-v1.5 es una extracción de características por BAAI.", "de": "bge-large-en-v1.5 ist eine Merkmalsextraktion von BAAI." } }, { "id": "Qwen/Qwen3-Embedding-8B", "name": "Qwen3-Embedding-8B", "author": "Qwen", "pipelineTag": "feature-extraction", "pipelineZh": "向量嵌入", "library": "sentence-transformers", "likes": 687, "downloads": 1571328, "trendingScore": 7, "createdAt": "2025-06-03T14:39:10.000Z", "lastModified": "2025-07-07T09:02:21.000Z", "tags": [ "sentence-transformers", "qwen3", "text-generation", "sentence-similarity", "feature-extraction", "text-embeddings-inference", "arxiv:2506.05176", "base_model:Qwen/Qwen3-8B-Base" ], "source": "HF Embedding Trending", "url": "https://huggingface.co/Qwen/Qwen3-Embedding-8B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3-Embedding-8B.png", "summary": "Qwen3-Embedding-8B 是面向文本嵌入与排序任务的大型模型，基于 Qwen3-8B 基座，具备多语言支持和卓越的文本检索与分类能力。", "title_i18n": { "en": "Qwen3-Embedding-8B", "zh-CN": "Qwen3-Embedding-8B", "ja": "Qwen3-Embedding-8B", "ko": "Qwen3-Embedding-8B", "es": "Qwen3-Embedding-8B", "de": "Qwen3-Embedding-8B" }, "summary_i18n": { "en": "Qwen3-Embedding-8B is a feature-extraction by Qwen.", "zh-CN": "Qwen3-Embedding-8B 是面向文本嵌入与排序任务的大型模型，基于 Qwen3-8B 基座，具备多语言支持和卓越的文本检索与分类能力。", "ja": "Qwen3-Embedding-8BはQwenによる特徴抽出モデルです。", "ko": "Qwen3-Embedding-8B은 Qwen의 특징 추출 모델입니다.", "es": "Qwen3-Embedding-8B es una extracción de características por Qwen.", "de": "Qwen3-Embedding-8B ist eine Merkmalsextraktion von Qwen." } }, { "id": "openvla/openvla-7b", "name": "openvla-7b", "author": "openvla", "pipelineTag": "robotics", "pipelineZh": "机器人", "library": "transformers", "likes": 219, "downloads": 1715742, "trendingScore": 1, "createdAt": "2024-06-10T16:35:59.000Z", "lastModified": "2026-02-17T03:43:23.000Z", "tags": [ "openvla", "feature-extraction", "robotics", "vla", "image-text-to-text", "multimodal", "pretraining", "en" ], "source": "HF Robotics Trending", "url": "https://huggingface.co/openvla/openvla-7b", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/openvla/openvla-7b.png", "summary": "OpenVLA-7B 是一个基于 97 万条机器人操作数据训练的开源视觉语言动作模型 支持多机器人控制和快速微调 具备强大的多模态理解和生成能力", "title_i18n": { "en": "openvla-7b", "zh-CN": "openvla-7b", "ja": "openvla-7b", "ko": "openvla-7b", "es": "openvla-7b", "de": "openvla-7b" }, "summary_i18n": { "en": "openvla-7b is a robotics by openvla.", "zh-CN": "OpenVLA-7B 是一个基于 97 万条机器人操作数据训练的开源视觉语言动作模型 支持多机器人控制和快速微调 具备强大的多模态理解和生成能力", "ja": "openvla-7b は openvla によるロボティクスです。", "ko": "openvla-7b는 openvla에서 개발한 로봇 기술입니다.", "es": "openvla-7b es un robot de openvla.", "de": "openvla-7b ist ein Roboter von openvla." } }, { "id": "nvidia/Alpamayo-1.5-10B", "name": "Alpamayo-1.5-10B", "author": "nvidia", "pipelineTag": "robotics", "pipelineZh": "机器人", "library": null, "likes": 70, "downloads": 79371, "trendingScore": 2, "createdAt": "2026-03-03T01:37:26.000Z", "lastModified": "2026-03-27T14:22:13.000Z", "tags": [ "alpamayo1_5", "robotics", "en", "dataset:nvidia/PhysicalAI-Autonomous-Vehicles", "dataset:nvidia/PhysicalAI-Autonomous-Vehicles-NuRec", "base_model:nvidia/Cosmos-Reason2-8B", "base_model:finetune:nvidia/Cosmos-Reason2-8B" ], "source": "HF Robotics Trending", "url": "https://huggingface.co/nvidia/Alpamayo-1.5-10B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/nvidia/Alpamayo-1.5-10B.png", "summary": "NVIDIA Alpamayo-1.5-10B 是一款面向自动驾驶的100亿参数视觉语言模型，基于 Cosmos-Reason2 优化，支持导航引导和多摄像头输入，具备强大的推理与决策能力。", "title_i18n": { "en": "Alpamayo-1.5-10B", "zh-CN": "Alpamayo-1.5-10B", "ja": "Alpamayo-1.5-10B", "ko": "Alpamayo-1.5-10B", "es": "Alpamayo-1.5-10B", "de": "Alpamayo-1.5-10B" }, "summary_i18n": { "en": "Alpamayo-1.5-10B is a robotics by nvidia.", "zh-CN": "NVIDIA Alpamayo-1.5-10B 是一款面向自动驾驶的100亿参数视觉语言模型，基于 Cosmos-Reason2 优化，支持导航引导和多摄像头输入，具备强大的推理与决策能力。", "ja": "Alpamayo-1.5-10B は nvidia によるロボティクスです。", "ko": "Alpamayo-1.5-10B는 nvidia에서 개발한 로봇 기술입니다.", "es": "Alpamayo-1.5-10B es un robot de nvidia.", "de": "Alpamayo-1.5-10B ist ein Roboter von nvidia." } }, { "id": "allenai/MolmoAct2-SO100_101", "name": "MolmoAct2-SO100_101", "author": "allenai", "pipelineTag": "robotics", "pipelineZh": "机器人", "library": "transformers", "likes": 10, "downloads": 1593, "trendingScore": 2, "createdAt": "2026-05-04T17:08:39.000Z", "lastModified": "2026-05-23T01:15:19.000Z", "tags": [ "molmoact2", "image-text-to-text", "robotics", "so100", "so101", "arxiv:2605.02881" ], "source": "HF Robotics Trending", "url": "https://huggingface.co/allenai/MolmoAct2-SO100_101", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/allenai/MolmoAct2-SO100_101.png", "summary": "MolmoAct2-SO100_101 是面向机器人控制的视觉语言动作模型，基于 Molmo2-ER 构建，支持连续动作预测，适用于 SO-100/101 任务推理与微调。", "title_i18n": { "en": "MolmoAct2-SO100_101", "zh-CN": "MolmoAct2-SO100_101", "ja": "MolmoAct2-SO100_101", "ko": "MolmoAct2-SO100_101", "es": "MolmoAct2-SO100_101", "de": "MolmoAct2-SO100_101" }, "summary_i18n": { "en": "MolmoAct2-SO100_101 is a robotics by allenai.", "zh-CN": "MolmoAct2-SO100_101 是面向机器人控制的视觉语言动作模型，基于 Molmo2-ER 构建，支持连续动作预测，适用于 SO-100/101 任务推理与微调。", "ja": "MolmoAct2-SO100_101 は allenai によるロボティクスです。", "ko": "MolmoAct2-SO100_101는 allenai에서 개발한 로봇 기술입니다.", "es": "MolmoAct2-SO100_101 es un robot de allenai.", "de": "MolmoAct2-SO100_101 ist ein Roboter von allenai." } }, { "id": "lerobot/smolvla_base", "name": "smolvla_base", "author": "lerobot", "pipelineTag": "robotics", "pipelineZh": "机器人", "library": "lerobot", "likes": 379, "downloads": 33355, "trendingScore": 2, "createdAt": "2025-06-01T07:51:59.000Z", "lastModified": "2026-01-22T09:32:49.000Z", "tags": [ "lerobot", "vision-language-action", "imitation-learning", "robotics", "en", "arxiv:2506.01844" ], "source": "HF Robotics Trending", "url": "https://huggingface.co/lerobot/smolvla_base", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/lerobot/smolvla_base.png", "summary": "SmolVLA 是一款面向低成本机器人的视觉-语言-动作模型，基于单块GPU训练，性能媲美大型模型，支持连续动作输出。", "title_i18n": { "en": "smolvla_base", "zh-CN": "smolvla_base", "ja": "smolvla_base", "ko": "smolvla_base", "es": "smolvla_base", "de": "smolvla_base" }, "summary_i18n": { "en": "smolvla_base is a robotics by lerobot.", "zh-CN": "SmolVLA 是一款面向低成本机器人的视觉-语言-动作模型，基于单块GPU训练，性能媲美大型模型，支持连续动作输出。", "ja": "smolvla_base は lerobot によるロボティクスです。", "ko": "smolvla_base는 lerobot에서 개발한 로봇 기술입니다.", "es": "smolvla_base es un robot de lerobot.", "de": "smolvla_base ist ein Roboter von lerobot." } }, { "id": "Supertone/supertonic-3", "name": "supertonic-3", "author": "Supertone", "pipelineTag": "text-to-speech", "pipelineZh": "语音合成", "library": "supertonic", "likes": 585, "downloads": 37545, "trendingScore": 328, "createdAt": "2026-05-06T20:46:20.000Z", "lastModified": "2026-05-18T08:59:01.000Z", "tags": [ "supertonic", "text-to-speech", "speech-synthesis", "tts", "multilingual", "on-device", "en", "ko" ], "source": "HF Trending", "url": "https://huggingface.co/Supertone/supertonic-3", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Supertone/supertonic-3.png", "summary": "Supertonic 3 是一款轻量级多语言文本转语音模型 支持31种语言 本地运行速度快且准确", "title_i18n": { "en": "supertonic-3", "zh-CN": "supertonic-3", "ja": "supertonic-3", "ko": "supertonic-3", "es": "supertonic-3", "de": "supertonic-3" }, "summary_i18n": { "en": "supertonic-3 is a text-to-speech by Supertone.", "zh-CN": "Supertonic 3 是一款轻量级多语言文本转语音模型 支持31种语言 本地运行速度快且准确", "ja": "supertonic-3 は Supertone によるテキストから音声への変換です。", "ko": "supertonic-3는 Supertone에서 개발한 텍스트-음성 변환 기술입니다.", "es": "supertonic-3 es una conversión texto-a-voz por Supertone.", "de": "supertonic-3 ist eine Text-zu-Sprache von Supertone." } }, { "id": "bytedance-research/Lance", "name": "Lance", "author": "bytedance-research", "pipelineTag": "any-to-any", "pipelineZh": "多模态(any2any)", "library": "Lance", "likes": 655, "downloads": 1001, "trendingScore": 631, "createdAt": "2026-05-15T08:05:57.000Z", "lastModified": "2026-05-22T16:55:56.000Z", "tags": [ "Lance", "multimodal", "image-generation", "video-generation", "image-editing", "video-understanding", "any-to-any", "arxiv:2605.18678" ], "source": "HF Trending", "url": "https://huggingface.co/bytedance-research/Lance", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/bytedance-research/Lance.png", "summary": "Lance 是由字节跳动研发的多模态统一模型支持图像视频生成与编辑等任务规模庞大能力全面覆盖多种媒体形式", "title_i18n": { "en": "Lance", "zh-CN": "Lance", "ja": "Lance", "ko": "Lance", "es": "Lance", "de": "Lance" }, "summary_i18n": { "en": "Lance is a any-to-any by bytedance-research.", "zh-CN": "Lance 是由字节跳动研发的多模态统一模型支持图像视频生成与编辑等任务规模庞大能力全面覆盖多种媒体形式", "ja": "Lance は bytedance-research によるエンドツーエンドのモデルです。", "ko": "Lance는 bytedance-research에서 개발한 다양한 유형의 모델입니다.", "es": "Lance es cualquier-a-cualquier por bytedance-research.", "de": "Lance ist ein Any-to-Any von bytedance-research." } }, { "id": "sapientinc/HRM-Text-1B", "name": "HRM-Text-1B", "author": "sapientinc", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 246, "downloads": 72470, "trendingScore": 243, "createdAt": "2026-05-17T15:13:02.000Z", "lastModified": "2026-05-21T05:57:38.000Z", "tags": [ "hrm_text", "text-generation", "hrm", "hierarchical-reasoning", "prefix-lm", "pre-alignment", "non-chat", "non-instruction-tuned" ], "source": "HF Trending", "url": "https://huggingface.co/sapientinc/HRM-Text-1B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/sapientinc/HRM-Text-1B.png", "summary": "基于HRM架构的1B参数语言模型，具备分层推理能力，适用于文本生成任务，适合进行进一步对齐优化。", "title_i18n": { "en": "HRM-Text-1B", "zh-CN": "HRM-Text-1B", "ja": "HRM-Text-1B", "ko": "HRM-Text-1B", "es": "HRM-Text-1B", "de": "HRM-Text-1B" }, "summary_i18n": { "en": "HRM-Text-1B is a text-generation by sapientinc.", "zh-CN": "基于HRM架构的1B参数语言模型，具备分层推理能力，适用于文本生成任务，适合进行进一步对齐优化。", "ja": "HRM-Text-1B は sapientinc によるテキスト生成です。", "ko": "HRM-Text-1B는 sapientinc에서 개발한 텍스트 생성 모델입니다.", "es": "HRM-Text-1B es generación de texto por sapientinc.", "de": "HRM-Text-1B ist eine Textgenerierung von sapientinc." } }, { "id": "openbmb/MiniCPM-V-4.6", "name": "MiniCPM-V-4.6", "author": "openbmb", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 906, "downloads": 221612, "trendingScore": 281, "createdAt": "2026-04-13T09:12:47.000Z", "lastModified": "2026-05-19T14:01:40.000Z", "tags": [ "minicpmv4_6", "image-text-to-text", "minicpm-v", "multimodal", "On-Device Model", "lightweight", "arxiv:2604.27393", "arxiv:2509.18154" ], "source": "HF Trending", "url": "https://huggingface.co/openbmb/MiniCPM-V-4.6", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/openbmb/MiniCPM-V-4.6.png", "summary": "面向手机端的轻量级多模态大模型支持图像和视频理解在计算效率和性能上表现优异", "title_i18n": { "en": "MiniCPM-V-4.6", "zh-CN": "MiniCPM-V-4.6", "ja": "MiniCPM-V-4.6", "ko": "MiniCPM-V-4.6", "es": "MiniCPM-V-4.6", "de": "MiniCPM-V-4.6" }, "summary_i18n": { "en": "MiniCPM-V-4.6 is a image-text-to-text by openbmb.", "zh-CN": "面向手机端的轻量级多模态大模型支持图像和视频理解在计算效率和性能上表现优异", "ja": "MiniCPM-V-4.6 は openbmb による画像・テキストからテキストへのモデルです。", "ko": "MiniCPM-V-4.6는 openbmb에서 개발한 이미지-텍스트-텍스트 모델입니다.", "es": "MiniCPM-V-4.6 es texto-imagen-a-texto por openbmb.", "de": "MiniCPM-V-4.6 ist eine Bild-Text-zu-Text von openbmb." } }, { "id": "Qwen/Qwen3.6-27B", "name": "Qwen3.6-27B", "author": "Qwen", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 1391, "downloads": 4049995, "trendingScore": 80, "createdAt": "2026-04-21T07:50:43.000Z", "lastModified": "2026-04-24T02:39:16.000Z", "tags": [ "qwen3_5", "image-text-to-text", "deploy:azure" ], "source": "HF Trending", "url": "https://huggingface.co/Qwen/Qwen3.6-27B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3.6-27B.png", "summary": "Qwen3.6-27B 是一款基于 270 亿参数的图像文本到文本模型具备强大的代码生成和多步骤任务处理能力适合开发高精度应用。", "title_i18n": { "en": "Qwen3.6-27B", "zh-CN": "Qwen3.6-27B", "ja": "Qwen3.6-27B", "ko": "Qwen3.6-27B", "es": "Qwen3.6-27B", "de": "Qwen3.6-27B" }, "summary_i18n": { "en": "Qwen3.6-27B is a image-text-to-text by Qwen.", "zh-CN": "Qwen3.6-27B 是一款基于 270 亿参数的图像文本到文本模型具备强大的代码生成和多步骤任务处理能力适合开发高精度应用。", "ja": "Qwen3.6-27BはQwenによる画像テキストからテキストへのモデルです。", "ko": "Qwen3.6-27B은 Qwen에서 만든 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-27B es un modelo de imagen-texto-a-texto por Qwen.", "de": "Qwen3.6-27B ist ein Bild-Text-zu-Text-Modell von Qwen." } }, { "id": "deepseek-ai/DeepSeek-V4-Flash", "name": "DeepSeek-V4-Flash", "author": "deepseek-ai", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 1194, "downloads": 2556531, "trendingScore": 84, "createdAt": "2026-04-22T06:04:20.000Z", "lastModified": "2026-05-06T04:18:09.000Z", "tags": [ "deepseek_v4", "text-generation", "8-bit", "fp8" ], "source": "HF Trending", "url": "https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/deepseek-ai/DeepSeek-V4-Flash.png", "summary": "面向文本生成的大型语言模型基于8位和fp8量化技术实现高效推理具备百万词上下文理解能力", "title_i18n": { "en": "DeepSeek-V4-Flash", "zh-CN": "DeepSeek-V4-Flash", "ja": "DeepSeek-V4-Flash", "ko": "DeepSeek-V4-Flash", "es": "DeepSeek-V4-Flash", "de": "DeepSeek-V4-Flash" }, "summary_i18n": { "en": "DeepSeek-V4-Flash is a text-generation by deepseek-ai.", "zh-CN": "面向文本生成的大型语言模型基于8位和fp8量化技术实现高效推理具备百万词上下文理解能力", "ja": "DeepSeek-V4-Flashはdeepseek-aiによるテキスト生成モデルです。", "ko": "DeepSeek-V4-Flash는 deepseek-ai에서 만든 텍스트 생성 모델입니다.", "es": "DeepSeek-V4-Flash es una generación de texto por deepseek-ai.", "de": "DeepSeek-V4-Flash ist eine Textgenerierung von deepseek-ai." } }, { "id": "NemoStation/Marlin-2B", "name": "Marlin-2B", "author": "NemoStation", "pipelineTag": "video-text-to-text", "pipelineZh": "video-text-to-text", "library": "transformers", "likes": 252, "downloads": 4002, "trendingScore": 245, "createdAt": "2026-05-13T16:23:12.000Z", "lastModified": "2026-05-20T07:54:57.000Z", "tags": [ "qwen3_5", "text-generation", "video", "multimodal", "video-captioning", "temporal-grounding", "qwen", "VLM" ], "source": "HF Trending", "url": "https://huggingface.co/NemoStation/Marlin-2B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/NemoStation/Marlin-2B.png", "summary": "定位视频到文本生成的大型语言模型 基于20亿参数规模 具备多模态理解和时序定位能力", "title_i18n": { "en": "Marlin-2B", "zh-CN": "Marlin-2B", "ja": "Marlin-2B", "ko": "Marlin-2B", "es": "Marlin-2B", "de": "Marlin-2B" }, "summary_i18n": { "en": "Marlin-2B is a video-text-to-text by NemoStation.", "zh-CN": "定位视频到文本生成的大型语言模型 基于20亿参数规模 具备多模态理解和时序定位能力", "ja": "Marlin-2BはNemoStationによる動画テキストからテキストへのモデルです。", "ko": "Marlin-2B는 NemoStation에서 만든 비디오-텍스트-텍스트 모델입니다.", "es": "Marlin-2B es un modelo de video-texto-a-texto por NemoStation.", "de": "Marlin-2B ist ein Video-Text-zu-Text-Modell von NemoStation." } }, { "id": "unsloth/Qwen3.6-35B-A3B-GGUF", "name": "Qwen3.6-35B-A3B-GGUF", "author": "unsloth", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 1096, "downloads": 2425597, "trendingScore": 44, "createdAt": "2026-04-16T04:31:31.000Z", "lastModified": "2026-04-20T12:42:25.000Z", "tags": [ "unsloth", "qwen", "qwen3_5_moe", "image-text-to-text", "base_model:Qwen/Qwen3.6-35B-A3B", "base_model:quantized:Qwen/Qwen3.6-35B-A3B", "deploy:azure", "imatrix" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/unsloth/Qwen3.6-35B-A3B-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/unsloth/Qwen3.6-35B-A3B-GGUF.png", "summary": "定位为图像文本到文本的生成模型，基于Qwen3.6-35B-A3B大模型，支持工具调用和代码生成能力。", "title_i18n": { "en": "Qwen3.6-35B-A3B-GGUF", "zh-CN": "Qwen3.6-35B-A3B-GGUF", "ja": "Qwen3.6-35B-A3B-GGUF", "ko": "Qwen3.6-35B-A3B-GGUF", "es": "Qwen3.6-35B-A3B-GGUF", "de": "Qwen3.6-35B-A3B-GGUF" }, "summary_i18n": { "en": "Qwen3.6-35B-A3B-GGUF is a image-text-to-text by unsloth.", "zh-CN": "定位为图像文本到文本的生成模型，基于Qwen3.6-35B-A3B大模型，支持工具调用和代码生成能力。", "ja": "Qwen3.6-35B-A3B-GGUFはunslothによる画像テキストからテキストへのモデルです。", "ko": "Qwen3.6-35B-A3B-GGUF는 unsloth에서 만든 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-35B-A3B-GGUF es un modelo de imagen-texto-a-texto por unsloth.", "de": "Qwen3.6-35B-A3B-GGUF ist ein Bild-Text-zu-Text-Modell von unsloth." } }, { "id": "DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "name": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "author": "DavidAU", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": null, "likes": 212, "downloads": 443744, "trendingScore": 46, "createdAt": "2026-04-29T02:37:13.000Z", "lastModified": "2026-05-14T00:27:54.000Z", "tags": [ "unsloth", "heretic", "uncensored", "abliterated", "fine tune", "creative", "creative writing", "fiction writing" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/DavidAU/Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF.png", "summary": "定位为图像文本到文本的生成模型 规模达27B参数 能力亮点包括超越原版性能、去审查及高效量化技术", "title_i18n": { "en": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "zh-CN": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "ja": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "ko": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "es": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF", "de": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF" }, "summary_i18n": { "en": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF is a image-text-to-text by DavidAU.", "zh-CN": "定位为图像文本到文本的生成模型 规模达27B参数 能力亮点包括超越原版性能、去审查及高效量化技术", "ja": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUFはDavidAUによる画像テキストからテキストへのモデルです。", "ko": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF는 DavidAU에서 만든 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF es un modelo de imagen-texto-a-texto por DavidAU.", "de": "Qwen3.6-27B-Heretic-Uncensored-FINETUNE-NEO-CODE-Di-IMatrix-MAX-GGUF ist ein Bild-Text-zu-Text-Modell von DavidAU." } }, { "id": "circlestone-labs/Anima", "name": "Anima", "author": "circlestone-labs", "pipelineTag": null, "pipelineZh": null, "library": "diffusion-single-file", "likes": 1500, "downloads": 602483, "trendingScore": 163, "createdAt": "2026-01-29T21:17:57.000Z", "lastModified": "2026-05-14T17:13:39.000Z", "tags": [ "diffusion-single-file", "comfyui" ], "source": "HF Trending", "url": "https://huggingface.co/circlestone-labs/Anima", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/circlestone-labs/Anima.png", "summary": "CircleStone Labs与Comfy Org联合开发的20亿参数文本生成动漫图像模型支持多种非写实艺术风格生成提供高灵活性和多样性", "title_i18n": { "en": "Anima", "zh-CN": "Anima", "ja": "Anima", "ko": "Anima", "es": "Anima", "de": "Anima" }, "summary_i18n": { "en": "Anima is a model by circlestone-labs.", "zh-CN": "CircleStone Labs与Comfy Org联合开发的20亿参数文本生成动漫图像模型支持多种非写实艺术风格生成提供高灵活性和多样性", "ja": "Animaはcirclestone-labsによるモデルです。", "ko": "Anima는 circlestone-labs에서 만든 모델입니다.", "es": "Anima es un modelo por circlestone-labs.", "de": "Anima ist ein Modell von circlestone-labs." } }, { "id": "Jackrong/Qwopus3.5-9B-Coder-GGUF", "name": "Qwopus3.5-9B-Coder-GGUF", "author": "Jackrong", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": "transformers", "likes": 164, "downloads": 28599, "trendingScore": 162, "createdAt": "2026-05-15T12:53:59.000Z", "lastModified": "2026-05-19T23:41:58.000Z", "tags": [ "unsloth", "qwen3_5", "reasoning", "chain-of-thought", "lora", "sft", "agent", "tool-use" ], "source": "HF Trending", "url": "https://huggingface.co/Jackrong/Qwopus3.5-9B-Coder-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Jackrong/Qwopus3.5-9B-Coder-GGUF.png", "summary": "基于Qwopus3.5-9B-v3.5优化，专攻编程与逻辑推理的开源模型，支持工具调用和视觉输入，具备高精度推理与代码处理能力。", "title_i18n": { "en": "Qwopus3.5-9B-Coder-GGUF", "zh-CN": "Qwopus3.5-9B-Coder-GGUF", "ja": "Qwopus3.5-9B-Coder-GGUF", "ko": "Qwopus3.5-9B-Coder-GGUF", "es": "Qwopus3.5-9B-Coder-GGUF", "de": "Qwopus3.5-9B-Coder-GGUF" }, "summary_i18n": { "en": "Qwopus3.5-9B-Coder-GGUF is a image-text-to-text by Jackrong.", "zh-CN": "基于Qwopus3.5-9B-v3.5优化，专攻编程与逻辑推理的开源模型，支持工具调用和视觉输入，具备高精度推理与代码处理能力。", "ja": "Qwopus3.5-9B-Coder-GGUFはJackrongによる画像テキストからテキストへのモデルです。", "ko": "Qwopus3.5-9B-Coder-GGUF는 Jackrong에서 만든 이미지-텍스트-텍스트 모델입니다.", "es": "Qwopus3.5-9B-Coder-GGUF es un modelo de imagen-texto-a-texto por Jackrong.", "de": "Qwopus3.5-9B-Coder-GGUF ist ein Bild-Text-zu-Text-Modell von Jackrong." } }, { "id": "tencent/Hy-MT2-1.8B", "name": "Hy-MT2-1.8B", "author": "tencent", "pipelineTag": "translation", "pipelineZh": "翻译", "library": "transformers", "likes": 281, "downloads": 564, "trendingScore": 273, "createdAt": "2026-05-11T08:22:10.000Z", "lastModified": "2026-05-22T05:09:24.000Z", "tags": [ "hunyuan_v1_dense", "text-generation", "translation", "zh", "en", "fr", "pt", "es" ], "source": "HF Trending", "url": "https://huggingface.co/tencent/Hy-MT2-1.8B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/tencent/Hy-MT2-1.8B.png", "summary": "腾讯Hy-MT2-1.8B是一款支持33种语言的多语言翻译模型具备强大的跨语言理解和生成能力同时通过极量化技术实现高效部署", "title_i18n": { "en": "Hy-MT2-1.8B", "zh-CN": "Hy-MT2-1.8B", "ja": "Hy-MT2-1.8B", "ko": "Hy-MT2-1.8B", "es": "Hy-MT2-1.8B", "de": "Hy-MT2-1.8B" }, "summary_i18n": { "en": "Hy-MT2-1.8B is a translation by tencent.", "zh-CN": "腾讯Hy-MT2-1.8B是一款支持33种语言的多语言翻译模型具备强大的跨语言理解和生成能力同时通过极量化技术实现高效部署", "ja": "Hy-MT2-1.8Bはtencentによる翻訳モデルです。", "ko": "Hy-MT2-1.8B는 tencent에서 만든 번역 모델입니다.", "es": "Hy-MT2-1.8B es una traducción por tencent.", "de": "Hy-MT2-1.8B ist eine Übersetzung von tencent." } }, { "id": "HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "name": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "author": "HauhauCS", "pipelineTag": "image-text-to-text", "pipelineZh": "图文问答 · 多模态", "library": null, "likes": 699, "downloads": 1218357, "trendingScore": 49, "createdAt": "2026-04-17T00:15:26.000Z", "lastModified": "2026-04-17T02:59:21.000Z", "tags": [ "uncensored", "qwen3.6", "moe", "vision", "multimodal", "image-text-to-text", "en", "zh" ], "source": "HF Multimodal Trending", "url": "https://huggingface.co/HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive.png", "summary": "基于Qwen3.6-35B大模型的图像文本到文本生成模型支持多模态输入具备强大无限制生成能力", "title_i18n": { "en": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "zh-CN": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "ja": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "ko": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "es": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive", "de": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive" }, "summary_i18n": { "en": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive is a image-text-to-text by HauhauCS.", "zh-CN": "基于Qwen3.6-35B大模型的图像文本到文本生成模型支持多模态输入具备强大无限制生成能力", "ja": "HauhauCSによって作成された画像テキストからテキストへのモデルです。", "ko": "HauhauCS의 이미지-텍스트-텍스트 모델입니다.", "es": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive es una imagen-texto-a-texto por HauhauCS.", "de": "Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive ist ein Bild-Text-zu-Text-Modell von HauhauCS." } }, { "id": "HiDream-ai/HiDream-O1-Image", "name": "HiDream-O1-Image", "author": "HiDream-ai", "pipelineTag": "image-text-to-image", "pipelineZh": "image-text-to-image", "library": "transformers", "likes": 422, "downloads": 22783, "trendingScore": 77, "createdAt": "2026-05-08T13:06:04.000Z", "lastModified": "2026-05-15T10:40:00.000Z", "tags": [ "qwen3_vl", "image-text-to-text", "image-text-to-image", "arxiv:2605.11061" ], "source": "HF Trending", "url": "https://huggingface.co/HiDream-ai/HiDream-O1-Image", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/HiDream-ai/HiDream-O1-Image.png", "summary": "HiDream-O1-Image 是一款基于像素级统一Transformer的图像生成大模型，支持文本到图像、图像编辑和主题个性化，最高生成2048×2048分辨率图像。", "title_i18n": { "en": "HiDream-O1-Image", "zh-CN": "HiDream-O1-Image", "ja": "HiDream-O1-Image", "ko": "HiDream-O1-Image", "es": "HiDream-O1-Image", "de": "HiDream-O1-Image" }, "summary_i18n": { "en": "HiDream-O1-Image is a image-text-to-image by HiDream-ai.", "zh-CN": "HiDream-O1-Image 是一款基于像素级统一Transformer的图像生成大模型，支持文本到图像、图像编辑和主题个性化，最高生成2048×2048分辨率图像。", "ja": "HiDream-aiによって作成された画像テキストから画像へのモデルです。", "ko": "HiDream-ai의 이미지-텍스트-이미지 모델입니다.", "es": "HiDream-O1-Image es una imagen-texto-a-imagen por HiDream-ai.", "de": "HiDream-O1-Image ist ein Bild-Text-zu-Bild-Modell von HiDream-ai." } }, { "id": "tencent/Hy-MT2-30B-A3B", "name": "Hy-MT2-30B-A3B", "author": "tencent", "pipelineTag": "translation", "pipelineZh": "翻译", "library": "transformers", "likes": 255, "downloads": 224, "trendingScore": 239, "createdAt": "2026-05-11T07:43:51.000Z", "lastModified": "2026-05-22T05:15:55.000Z", "tags": [ "hy_v3", "text-generation", "translation", "zh", "en", "fr", "pt", "es" ], "source": "HF Trending", "url": "https://huggingface.co/tencent/Hy-MT2-30B-A3B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/tencent/Hy-MT2-30B-A3B.png", "summary": "腾讯Hy-MT2-30B-A3B是一款30B参数的多语言翻译模型支持33种语言翻译具备强大的指令遵循能力和多场景翻译性能优于同类开源模型和部分商业API。", "title_i18n": { "en": "Hy-MT2-30B-A3B", "zh-CN": "Hy-MT2-30B-A3B", "ja": "Hy-MT2-30B-A3B", "ko": "Hy-MT2-30B-A3B", "es": "Hy-MT2-30B-A3B", "de": "Hy-MT2-30B-A3B" }, "summary_i18n": { "en": "Hy-MT2-30B-A3B is a translation by tencent.", "zh-CN": "腾讯Hy-MT2-30B-A3B是一款30B参数的多语言翻译模型支持33种语言翻译具备强大的指令遵循能力和多场景翻译性能优于同类开源模型和部分商业API。", "ja": "tencentによって作成された翻訳モデルです。", "ko": "tencent의 번역 모델입니다.", "es": "Hy-MT2-30B-A3B es una traducción por tencent.", "de": "Hy-MT2-30B-A3B ist eine Übersetzung von tencent." } }, { "id": "Jiunsong/supergemma4-26b-uncensored-gguf-v2", "name": "supergemma4-26b-uncensored-gguf-v2", "author": "Jiunsong", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": null, "likes": 657, "downloads": 234418, "trendingScore": 47, "createdAt": "2026-04-11T11:03:05.000Z", "lastModified": "2026-04-12T13:42:43.000Z", "tags": [ "gemma4", "uncensored", "fast", "llama.cpp", "apple-silicon", "korean", "coding", "tool-use" ], "source": "HF LLM Trending", "url": "https://huggingface.co/Jiunsong/supergemma4-26b-uncensored-gguf-v2", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Jiunsong/supergemma4-26b-uncensored-gguf-v2.png", "summary": "定位为高性能本地文本生成模型，基于Gemma4-26B基座，支持韩语、代码和工具调用，具备快速推理与去中心化聊天特性。", "title_i18n": { "en": "supergemma4-26b-uncensored-gguf-v2", "zh-CN": "supergemma4-26b-uncensored-gguf-v2", "ja": "supergemma4-26b-uncensored-gguf-v2", "ko": "supergemma4-26b-uncensored-gguf-v2", "es": "supergemma4-26b-uncensored-gguf-v2", "de": "supergemma4-26b-uncensored-gguf-v2" }, "summary_i18n": { "en": "supergemma4-26b-uncensored-gguf-v2 is a text-generation by Jiunsong.", "zh-CN": "定位为高性能本地文本生成模型，基于Gemma4-26B基座，支持韩语、代码和工具调用，具备快速推理与去中心化聊天特性。", "ja": "Jiunsongによって作成されたテキスト生成モデルです。", "ko": "Jiunsong의 텍스트 생성 모델입니다.", "es": "supergemma4-26b-uncensored-gguf-v2 es una generación de texto por Jiunsong.", "de": "supergemma4-26b-uncensored-gguf-v2 ist eine Textgenerierung von Jiunsong." } }, { "id": "OBLITERATUS/gemma-4-E4B-it-OBLITERATED", "name": "gemma-4-E4B-it-OBLITERATED", "author": "OBLITERATUS", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": null, "likes": 647, "downloads": 359608, "trendingScore": 34, "createdAt": "2026-04-15T01:22:42.000Z", "lastModified": "2026-04-19T21:43:06.000Z", "tags": [ "gemma4", "abliterated", "uncensored", "obliteratus", "refusal-removal", "text-generation", "base_model:google/gemma-4-E4B-it", "base_model:quantized:google/gemma-4-E4B-it" ], "source": "HF LLM Trending", "url": "https://huggingface.co/OBLITERATUS/gemma-4-E4B-it-OBLITERATED", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/OBLITERATUS/gemma-4-E4B-it-OBLITERATED.png", "summary": "OBLITERATUS/gemma-4-E4B-it-OBLITERATED 是基于 Gemma 4 的文本生成模型 去除了所有限制 具备强大生成能力 可在手机端运行", "title_i18n": { "en": "gemma-4-E4B-it-OBLITERATED", "zh-CN": "gemma-4-E4B-it-OBLITERATED", "ja": "gemma-4-E4B-it-OBLITERATED", "ko": "gemma-4-E4B-it-OBLITERATED", "es": "gemma-4-E4B-it-OBLITERATED", "de": "gemma-4-E4B-it-OBLITERATED" }, "summary_i18n": { "en": "gemma-4-E4B-it-OBLITERATED is a text-generation by OBLITERATUS.", "zh-CN": "OBLITERATUS/gemma-4-E4B-it-OBLITERATED 是基于 Gemma 4 的文本生成模型 去除了所有限制 具备强大生成能力 可在手机端运行", "ja": "OBLITERATUSによって作成されたテキスト生成モデルです。", "ko": "OBLITERATUS의 텍스트 생성 모델입니다.", "es": "gemma-4-E4B-it-OBLITERATED es una generación de texto por OBLITERATUS.", "de": "gemma-4-E4B-it-OBLITERATED ist eine Textgenerierung von OBLITERATUS." } }, { "id": "Jackrong/Qwopus3.5-9B-Coder-MTP-GGUF", "name": "Qwopus3.5-9B-Coder-MTP-GGUF", "author": "Jackrong", "pipelineTag": "text-generation", "pipelineZh": "文本生成 · LLM", "library": "transformers", "likes": 69, "downloads": 21448, "trendingScore": 66, "createdAt": "2026-05-18T05:48:32.000Z", "lastModified": "2026-05-19T23:43:59.000Z", "tags": [ "unsloth", "qwen3_5", "reasoning", "chain-of-thought", "mtp", "multi-token-prediction", "speculative-decoding", "lora" ], "source": "HF LLM Trending", "url": "https://huggingface.co/Jackrong/Qwopus3.5-9B-Coder-MTP-GGUF", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Jackrong/Qwopus3.5-9B-Coder-MTP-GGUF.png", "summary": "定位为代码生成与复杂推理的大型语言模型 规模为9B参数 基于MTP技术实现多token预测和推理加速", "title_i18n": { "en": "Qwopus3.5-9B-Coder-MTP-GGUF", "zh-CN": "Qwopus3.5-9B-Coder-MTP-GGUF", "ja": "Qwopus3.5-9B-Coder-MTP-GGUF", "ko": "Qwopus3.5-9B-Coder-MTP-GGUF", "es": "Qwopus3.5-9B-Coder-MTP-GGUF", "de": "Qwopus3.5-9B-Coder-MTP-GGUF" }, "summary_i18n": { "en": "Qwopus3.5-9B-Coder-MTP-GGUF is a text-generation by Jackrong.", "zh-CN": "定位为代码生成与复杂推理的大型语言模型 规模为9B参数 基于MTP技术实现多token预测和推理加速", "ja": "Jackrongによって作成されたテキスト生成モデルです。", "ko": "Jackrong의 텍스트 생성 모델입니다.", "es": "Qwopus3.5-9B-Coder-MTP-GGUF es una generación de texto por Jackrong.", "de": "Qwopus3.5-9B-Coder-MTP-GGUF ist eine Textgenerierung von Jackrong." } }, { "id": "numind/NuExtract3", "name": "NuExtract3", "author": "numind", "pipelineTag": "image-to-text", "pipelineZh": "image-to-text", "library": "transformers", "likes": 78, "downloads": 7576, "trendingScore": 76, "createdAt": "2026-04-29T07:46:41.000Z", "lastModified": "2026-05-20T09:24:47.000Z", "tags": [ "qwen3_5", "image-text-to-text", "vision-language", "vlm", "document-understanding", "structured-extraction", "information-extraction", "ocr" ], "source": "HF Trending", "url": "https://huggingface.co/numind/NuExtract3", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/numind/NuExtract3.png", "summary": "NuExtract3 是一款 4B 规模的视觉语言模型专注于文档理解与结构化信息提取支持多模态输入和高质量图像转 Markdown 能力适用于 OCR 和 RAG 预处理等场景。", "title_i18n": { "en": "NuExtract3", "zh-CN": "NuExtract3", "ja": "NuExtract3", "ko": "NuExtract3", "es": "NuExtract3", "de": "NuExtract3" }, "summary_i18n": { "en": "NuExtract3 is a image-to-text by numind.", "zh-CN": "NuExtract3 是一款 4B 规模的视觉语言模型专注于文档理解与结构化信息提取支持多模态输入和高质量图像转 Markdown 能力适用于 OCR 和 RAG 预处理等场景。", "ja": "numindによって作成された画像からテキストへのモデルです。", "ko": "numind의 이미지-텍스트 모델입니다.", "es": "NuExtract3 es una imagen-texto por numind.", "de": "NuExtract3 ist ein Bild-zu-Text-Modell von numind." } }, { "id": "black-forest-labs/FLUX.1-schnell", "name": "FLUX.1-schnell", "author": "black-forest-labs", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 4905, "downloads": 622869, "trendingScore": 14, "createdAt": "2024-07-31T19:58:05.000Z", "lastModified": "2024-08-16T14:37:56.000Z", "tags": [ "text-to-image", "image-generation", "flux", "en", "deploy:azure" ], "source": "HF Image Trending", "url": "https://huggingface.co/black-forest-labs/FLUX.1-schnell", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/black-forest-labs/FLUX.1-schnell.png", "summary": "生成高质量图像的文本到图像模型 基于FLUX架构 具备快速生成和细腻细节表现能力", "title_i18n": { "en": "FLUX.1-schnell", "zh-CN": "FLUX.1-schnell", "ja": "FLUX.1-schnell", "ko": "FLUX.1-schnell", "es": "FLUX.1-schnell", "de": "FLUX.1-schnell" }, "summary_i18n": { "en": "FLUX.1-schnell is a text-to-image by black-forest-labs.", "zh-CN": "生成高质量图像的文本到图像模型 基于FLUX架构 具备快速生成和细腻细节表现能力", "ja": "black-forest-labsによって作成されたテキストから画像へのモデルです。", "ko": "black-forest-labs의 텍스트-이미지 모델입니다.", "es": "FLUX.1-schnell es una texto-a-imagen por black-forest-labs.", "de": "FLUX.1-schnell ist ein Text-zu-Bild-Modell von black-forest-labs." } }, { "id": "RunDiffusion/Juggernaut-Z-Image", "name": "Juggernaut-Z-Image", "author": "RunDiffusion", "pipelineTag": "text-to-image", "pipelineZh": "文生图", "library": "diffusers", "likes": 94, "downloads": 21645, "trendingScore": 26, "createdAt": "2026-04-29T17:33:22.000Z", "lastModified": "2026-05-13T18:41:56.000Z", "tags": [ "text-to-image", "rundiffusion", "z-image", "en", "base_model:Tongyi-MAI/Z-Image", "base_model:quantized:Tongyi-MAI/Z-Image" ], "source": "HF Image Trending", "url": "https://huggingface.co/RunDiffusion/Juggernaut-Z-Image", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/RunDiffusion/Juggernaut-Z-Image.png", "summary": "Juggernaut-Z-Image 是基于 Z-Image 基座的文本生成图像模型，专注于提升光影效果与画面质感，具备 cinematic 风格的高质量输出能力。", "title_i18n": { "en": "Juggernaut-Z-Image", "zh-CN": "Juggernaut-Z-Image", "ja": "Juggernaut-Z-Image", "ko": "Juggernaut-Z-Image", "es": "Juggernaut-Z-Image", "de": "Juggernaut-Z-Image" }, "summary_i18n": { "en": "Juggernaut-Z-Image is a text-to-image by RunDiffusion.", "zh-CN": "Juggernaut-Z-Image 是基于 Z-Image 基座的文本生成图像模型，专注于提升光影效果与画面质感，具备 cinematic 风格的高质量输出能力。", "ja": "Juggernaut-Z-ImageはRunDiffusionによるテキストから画像生成のモデルです。", "ko": "Juggernaut-Z-Image는 RunDiffusion의 텍스트-이미지 모델입니다.", "es": "Juggernaut-Z-Image es una generación de imágenes a partir de texto por RunDiffusion.", "de": "Juggernaut-Z-Image ist ein Text-zu-Bild-Modell von RunDiffusion." } }, { "id": "Qwen/Qwen3-ASR-1.7B", "name": "Qwen3-ASR-1.7B", "author": "Qwen", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": null, "likes": 819, "downloads": 1975558, "trendingScore": 13, "createdAt": "2026-01-28T03:22:40.000Z", "lastModified": "2026-01-30T03:00:12.000Z", "tags": [ "qwen3_asr", "automatic-speech-recognition", "arxiv:2601.21337", "deploy:azure" ], "source": "HF Audio Trending", "url": "https://huggingface.co/Qwen/Qwen3-ASR-1.7B", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/Qwen/Qwen3-ASR-1.7B.png", "summary": "Qwen3-ASR-1.7B 是支持 52 种语言和方言的自动语音识别模型基于 Qwen3-Omni 大模型具备高精度和快速推理能力适用于复杂声学环境和长音频处理。", "title_i18n": { "en": "Qwen3-ASR-1.7B", "zh-CN": "Qwen3-ASR-1.7B", "ja": "Qwen3-ASR-1.7B", "ko": "Qwen3-ASR-1.7B", "es": "Qwen3-ASR-1.7B", "de": "Qwen3-ASR-1.7B" }, "summary_i18n": { "en": "Qwen3-ASR-1.7B is a automatic-speech-recognition by Qwen.", "zh-CN": "Qwen3-ASR-1.7B 是支持 52 种语言和方言的自动语音识别模型基于 Qwen3-Omni 大模型具备高精度和快速推理能力适用于复杂声学环境和长音频处理。", "ja": "Qwen3-ASR-1.7BはQwenによる音声認識のモデルです。", "ko": "Qwen3-ASR-1.7B는 Qwen의 자동 음성 인식 모델입니다.", "es": "Qwen3-ASR-1.7B es un reconocimiento de voz automático por Qwen.", "de": "Qwen3-ASR-1.7B ist eine Spracherkennung von Qwen." } }, { "id": "pyannote/speaker-diarization-community-1", "name": "speaker-diarization-community-1", "author": "pyannote", "pipelineTag": "automatic-speech-recognition", "pipelineZh": "语音识别", "library": "pyannote-audio", "likes": 395, "downloads": 2912607, "trendingScore": 17, "createdAt": "2025-04-15T21:31:35.000Z", "lastModified": "2025-09-29T08:43:24.000Z", "tags": [ "pyannote-audio", "pyannote", "pyannote-audio-pipeline", "audio", "voice", "speech", "speaker", "speaker-diarization" ], "source": "HF Audio Trending", "url": "https://huggingface.co/pyannote/speaker-diarization-community-1", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/pyannote/speaker-diarization-community-1.png", "summary": "定位为说话人分割模型 规模基于大型音频数据集 具备高精度的语音识别与说话人区分能力", "title_i18n": { "en": "speaker-diarization-community-1", "zh-CN": "speaker-diarization-community-1", "ja": "speaker-diarization-community-1", "ko": "speaker-diarization-community-1", "es": "speaker-diarization-community-1", "de": "speaker-diarization-community-1" }, "summary_i18n": { "en": "speaker-diarization-community-1 is a automatic-speech-recognition by pyannote.", "zh-CN": "定位为说话人分割模型 规模基于大型音频数据集 具备高精度的语音识别与说话人区分能力", "ja": "speaker-diarization-community-1はpyannoteによる音声認識のモデルです。", "ko": "speaker-diarization-community-1은 pyannote의 자동 음성 인식 모델입니다.", "es": "speaker-diarization-community-1 es un reconocimiento de voz automático por pyannote.", "de": "speaker-diarization-community-1 ist eine Spracherkennung von pyannote." } }, { "id": "ResembleAI/Dramabox", "name": "Dramabox", "author": "ResembleAI", "pipelineTag": "text-to-speech", "pipelineZh": "语音合成", "library": "ltx-audio-tts", "likes": 233, "downloads": 1354, "trendingScore": 137, "createdAt": "2026-04-17T07:17:52.000Z", "lastModified": "2026-05-13T16:39:38.000Z", "tags": [ "ltx-audio-tts", "dramabox-tts", "tts", "voice-cloning", "audio-generation", "diffusion-transformer", "flow-matching", "ltx-2" ], "source": "HF Trending", "url": "https://huggingface.co/ResembleAI/Dramabox", "thumbnail": "https://cdn-thumbnails.huggingface.co/social-thumbnails/models/ResembleAI/Dramabox.png", "summary": "ResembleAI/Dramabox 是基于 LTX-2.3 3.3B 模型的语音合成系统 支持通过提示词控制说话人身份、情感和表达细节 具备语音克隆能力", "title_i18n": { "en": "Dramabox", "zh-CN": "Dramabox", "ja": "Dramabox", "ko": "Dramabox", "es": "Dramabox", "de": "Dramabox" }, "summary_i18n": { "en": "Dramabox is a text-to-speech by ResembleAI.", "zh-CN": "ResembleAI/Dramabox 是基于 LTX-2.3 3.3B 模型的语音合成系统 支持通过提示词控制说话人身份、情感和表达细节 具备语音克隆能力", "ja": "DramaboxはResembleAIによるテキストから音声生成のモデルです。", "ko": "Dramabox는 ResembleAI의 텍스트-음성 모델입니다.", "es": "Dramabox es una generación de voz a partir de texto por ResembleAI.", "de": "Dramabox ist ein Text-zu-Sprache-Modell von ResembleAI." } } ], "i18nUpdatedAt": "2026-05-23T05:47:47.119Z", "locales": [ "en", "zh-CN", "ja", "ko", "es", "de" ] }