@huggingface/tasks 0.11.7 → 0.11.9

package/dist/index.cjs

@@ -1388,6 +1388,12 @@ var PIPELINE_DATA = {
     modality: "cv",
     color: "indigo"
   },
+  "video-text-to-text": {
+    name: "Video-Text-to-Text",
+    modality: "multimodal",
+    color: "blue",
+    hideInDatasets: true
+  },
   other: {
     name: "Other",
     modality: "other",
@@ -1731,8 +1737,12 @@ var taskData5 = {
   ],
   spaces: [
     {
-      description: "A leaderboard to rank best feature extraction models..",
+      description: "A leaderboard to rank text feature extraction models based on a benchmark.",
       id: "mteb/leaderboard"
+    },
+    {
+      description: "A leaderboard to rank best feature extraction models based on human feedback.",
+      id: "mteb/arena"
     }
   ],
   summary: "Feature extraction is the task of extracting features learnt in a model.",
@@ -1937,15 +1947,19 @@ var taskData8 = {
     },
     {
       description: "A strong image feature extraction model.",
-      id: "google/vit-base-patch16-224-in21k"
+      id: "nvidia/MambaVision-T-1K"
     },
     {
-      description: "A robust image feature extraction models.",
+      description: "A robust image feature extraction model.",
       id: "facebook/dino-vitb16"
     },
     {
-      description: "Strong image-text-to-text model made for information retrieval from documents.",
+      description: "Strong image feature extraction model made for information retrieval from documents.",
       id: "vidore/colpali"
+    },
+    {
+      description: "Strong image feature extraction model that can be used on images and documents.",
+      id: "OpenGVLab/InternViT-6B-448px-V1-2"
     }
   ],
   spaces: [],
@@ -1997,8 +2011,8 @@ var taskData9 = {
   ],
   models: [
     {
-      description: "A model that enhances images captured in low light conditions.",
-      id: "keras-io/low-light-image-enhancement"
+      description: "An image-to-image model to improve image resolution.",
+      id: "fal/AuraSR-v2"
     },
     {
       description: "A model that increases the resolution of an image.",
@@ -2216,7 +2230,7 @@ var taskData11 = {
   ],
   summary: "Image-text-to-text models take in an image and text prompt and output text. These models are also called vision-language models, or VLMs. The difference from image-to-text models is that these models take an additional text input, not restricting the model to certain use cases like image captioning, and may also be trained to accept a conversation as input.",
   widgetModels: ["microsoft/kosmos-2-patch14-224"],
-  youtubeId: ""
+  youtubeId: "IoGaGfU1CIg"
 };
 var data_default11 = taskData11;
 
@@ -2267,16 +2281,16 @@ var taskData12 = {
       id: "facebook/detr-resnet-50-panoptic"
     },
     {
-      description: "Semantic segmentation model trained on ADE20k benchmark dataset.",
-      id: "microsoft/beit-large-finetuned-ade-640-640"
+      description: "Background removal model.",
+      id: "briaai/RMBG-1.4"
     },
     {
       description: "Semantic segmentation model trained on ADE20k benchmark dataset with 512x512 resolution.",
       id: "nvidia/segformer-b0-finetuned-ade-512-512"
     },
     {
-      description: "Semantic segmentation model trained Cityscapes dataset.",
-      id: "facebook/mask2former-swin-large-cityscapes-semantic"
+      description: "A multipurpose image segmentation model for high resolution images.",
+      id: "ZhengPeng7/BiRefNet"
     },
     {
       description: "Panoptic segmentation model trained COCO (common objects) dataset.",
@@ -2340,13 +2354,13 @@ var taskData13 = {
     },
     {
       description: "Very strong mask generation model.",
-      id: "facebook/sam-vit-huge"
+      id: "facebook/sam2-hiera-large"
     }
   ],
   spaces: [
     {
-      description: "An application that combines a mask generation model with an image embedding model for open-vocabulary image segmentation.",
-      id: "SkalskiP/SAM_and_MetaCLIP"
+      description: "An application that combines a mask generation model with a zero-shot object detection model for text-guided image segmentation.",
+      id: "merve/OWLSAM2"
     },
     {
       description: "An application that compares the performance of a large and a small mask generation model.",
@@ -2354,7 +2368,7 @@ var taskData13 = {
     },
     {
       description: "An application based on an improved mask generation model.",
-      id: "linfanluntan/Grounded-SAM"
+      id: "SkalskiP/segment-anything-model-2"
     },
     {
       description: "An application to remove objects from videos using mask generation models.",
@@ -3050,15 +3064,15 @@ var taskData24 = {
   models: [
     {
       description: "One of the most powerful image generation models that can generate realistic outputs.",
-      id: "stabilityai/stable-diffusion-xl-base-1.0"
+      id: "black-forest-labs/FLUX.1-dev"
     },
     {
       description: "A powerful yet fast image generation model.",
       id: "latent-consistency/lcm-lora-sdxl"
     },
     {
-      description: "A very fast text-to-image model.",
-      id: "ByteDance/SDXL-Lightning"
+      description: "Text-to-image model for photorealistic generation.",
+      id: "Kwai-Kolors/Kolors"
     },
     {
       description: "A powerful text-to-image model.",
@@ -3419,6 +3433,10 @@ var taskData29 = {
     {
       description: "An instruction dataset with preference ratings on responses.",
       id: "openbmb/UltraFeedback"
+    },
+    {
+      description: "A large synthetic dataset for alignment of text generation models.",
+      id: "argilla/magpie-ultra-v0.1"
     }
   ],
   demo: {
@@ -3449,32 +3467,32 @@ var taskData29 = {
   ],
   models: [
     {
-      description: "A large language model trained for text generation.",
-      id: "bigscience/bloom-560m"
+      description: "A text-generation model trained to follow instructions.",
+      id: "google/gemma-2-2b-it"
     },
     {
-      description: "A large code generation model that can generate code in 80+ languages.",
+      description: "A code generation model that can generate code in 80+ languages.",
       id: "bigcode/starcoder"
     },
     {
-      description: "A very powerful text generation model.",
-      id: "mistralai/Mixtral-8x7B-Instruct-v0.1"
+      description: "Very powerful text generation model trained to follow instructions.",
+      id: "meta-llama/Meta-Llama-3.1-8B-Instruct"
     },
     {
       description: "Small yet powerful text generation model.",
-      id: "microsoft/phi-2"
+      id: "microsoft/Phi-3-mini-4k-instruct"
     },
     {
-      description: "A very powerful model that can chat, do mathematical reasoning and write code.",
-      id: "openchat/openchat-3.5-0106"
+      description: "A very powerful model that can solve mathematical problems.",
+      id: "AI-MO/NuminaMath-7B-TIR"
     },
     {
-      description: "Very strong yet small assistant model.",
-      id: "HuggingFaceH4/zephyr-7b-beta"
+      description: "Strong coding assistant model.",
+      id: "HuggingFaceH4/starchat2-15b-v0.1"
     },
     {
       description: "Very strong open-source large language model.",
-      id: "meta-llama/Llama-2-70b-hf"
+      id: "mistralai/Mistral-Nemo-Instruct-2407"
     }
   ],
   spaces: [
@@ -3501,7 +3519,7 @@ var taskData29 = {
   ],
   summary: "Generating text is the task of generating new text given another text. These models can, for example, fill in incomplete text or paraphrase.",
   widgetModels: ["HuggingFaceH4/zephyr-7b-beta"],
-  youtubeId: "Vpjb1lu0MDk"
+  youtubeId: "e9gNEAlsOvU"
 };
 var data_default29 = taskData29;
 
@@ -4226,6 +4244,7 @@ var TASKS_MODEL_LIBRARIES = {
   ],
   translation: ["transformers", "transformers.js"],
   "unconditional-image-generation": ["diffusers"],
+  "video-text-to-text": ["transformers"],
   "visual-question-answering": ["transformers", "transformers.js"],
   "voice-activity-detection": [],
   "zero-shot-classification": ["transformers", "transformers.js"],
@@ -4285,6 +4304,7 @@ var TASKS_DATA = {
   "token-classification": getData("token-classification", data_default26),
   translation: getData("translation", data_default27),
   "unconditional-image-generation": getData("unconditional-image-generation", data_default31),
+  "video-text-to-text": getData("video-text-to-text", data_default16),
   "visual-question-answering": getData("visual-question-answering", data_default33),
   "voice-activity-detection": void 0,
   "zero-shot-classification": getData("zero-shot-classification", data_default34),
@@ -4522,6 +4542,23 @@ tokenizer = keras_nlp.models.Tokenizer.from_preset("hf://${model.id}")
 backbone = keras_nlp.models.Backbone.from_preset("hf://${model.id}")
 `
 ];
+var llama_cpp_python = (model) => [
+  `from llama_cpp import Llama
+
+llm = Llama.from_pretrained(
+	repo_id="${model.id}",
+	filename="{{GGUF_FILE}}",
+)
+
+llm.create_chat_completion(
+		messages = [
+			{
+				"role": "user",
+				"content": "What is the capital of France?"
+			}
+		]
+)`
+];
 var tf_keras = (model) => [
   `# Note: 'keras<3.x' or 'tf_keras' must be installed (legacy)
 # See https://github.com/keras-team/tf-keras for more details.
@@ -4747,6 +4784,33 @@ var fastai = (model) => [
 
 learn = from_pretrained_fastai("${model.id}")`
 ];
+var sam2 = (model) => {
+  const image_predictor = `# Use SAM2 with images
+import torch
+from sam2.sam2_image_predictor import SAM2ImagePredictor
+
+predictor = SAM2ImagePredictor.from_pretrained(${model.id})
+
+with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
+    predictor.set_image(<your_image>)
+    masks, _, _ = predictor.predict(<input_prompts>)`;
+  const video_predictor = `# Use SAM2 with videos
+import torch
+from sam2.sam2_video_predictor import SAM2VideoPredictor
+	
+predictor = SAM2VideoPredictor.from_pretrained(${model.id})
+
+with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
+    state = predictor.init_state(<your_video>)
+
+    # add new prompts and instantly get the output on the same frame
+    frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>):
+
+    # propagate the prompts to get masklets throughout the video
+    for frame_idx, object_ids, masks in predictor.propagate_in_video(state):
+        ...`;
+  return [image_predictor, video_predictor];
+};
 var sampleFactory = (model) => [
   `python -m sample_factory.huggingface.load_from_hub -r ${model.id} -d ./train_dir`
 ];
@@ -5292,6 +5356,12 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
     filter: false,
     countDownloads: `path:"liveportrait/landmark.onnx"`
   },
+  "llama-cpp-python": {
+    prettyLabel: "llama-cpp-python",
+    repoName: "llama-cpp-python",
+    repoUrl: "https://github.com/abetlen/llama-cpp-python",
+    snippets: llama_cpp_python
+  },
   mindspore: {
     prettyLabel: "MindSpore",
     repoName: "mindspore",
@@ -5407,6 +5477,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
     filter: false,
     countDownloads: `path:"tokenizer.model"`
   },
+  refiners: {
+    prettyLabel: "Refiners",
+    repoName: "Refiners",
+    repoUrl: "https://github.com/finegrain-ai/refiners",
+    docsUrl: "https://refine.rs/",
+    filter: false,
+    countDownloads: `path:"model.safetensors"`
+  },
   saelens: {
     prettyLabel: "SAELens",
     repoName: "SAELens",
@@ -5414,6 +5492,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
     snippets: saelens,
     filter: false
   },
+  sam2: {
+    prettyLabel: "sam2",
+    repoName: "sam2",
+    repoUrl: "https://github.com/facebookresearch/segment-anything-2",
+    filter: false,
+    snippets: sam2,
+    countDownloads: `path_extension:"pt"`
+  },
   "sample-factory": {
     prettyLabel: "sample-factory",
     repoName: "sample-factory",
@@ -6555,6 +6641,27 @@ var snippetLlamacpp = (model, filepath) => {
     }
   ];
 };
+var snippetLocalAI = (model, filepath) => {
+  const command = (binary) => ["# Load and run the model:", `${binary} huggingface://${model.id}/${filepath ?? "{{GGUF_FILE}}"}`].join("\n");
+  return [
+    {
+      title: "Install from binary",
+      setup: "curl https://localai.io/install.sh | sh",
+      content: command("local-ai run")
+    },
+    {
+      title: "Use Docker images",
+      setup: [
+        // prettier-ignore
+        "# Pull the image:",
+        "docker pull localai/localai:latest-cpu"
+      ].join("\n"),
+      content: command(
+        "docker run -p 8080:8080 --name localai -v $PWD/models:/build/models localai/localai:latest-cpu"
+      )
+    }
+  ];
+};
 var LOCAL_APPS = {
   "llama.cpp": {
     prettyLabel: "llama.cpp",
@@ -6570,6 +6677,13 @@ var LOCAL_APPS = {
     displayOnModelPage: isGgufModel,
     deeplink: (model, filepath) => new URL(`lmstudio://open_from_hf?model=${model.id}${filepath ? `&file=${filepath}` : ""}`)
   },
+  localai: {
+    prettyLabel: "LocalAI",
+    docsUrl: "https://github.com/mudler/LocalAI",
+    mainTask: "text-generation",
+    displayOnModelPage: isGgufModel,
+    snippet: snippetLocalAI
+  },
   jan: {
     prettyLabel: "Jan",
     docsUrl: "https://jan.ai",
@@ -6640,9 +6754,23 @@ var LOCAL_APPS = {
     docsUrl: "https://diffusionbee.com",
     mainTask: "text-to-image",
     macOSOnly: true,
-    comingSoon: true,
     displayOnModelPage: (model) => model.library_name === "diffusers" && model.pipeline_tag === "text-to-image",
     deeplink: (model) => new URL(`diffusionbee://open_from_hf?model=${model.id}`)
+  },
+  joyfusion: {
+    prettyLabel: "JoyFusion",
+    docsUrl: "https://joyfusion.app",
+    mainTask: "text-to-image",
+    macOSOnly: true,
+    displayOnModelPage: (model) => model.tags.includes("coreml") && model.pipeline_tag === "text-to-image",
+    deeplink: (model) => new URL(`https://joyfusion.app/import_from_hf?repo_id=${model.id}`)
+  },
+  invoke: {
+    prettyLabel: "Invoke",
+    docsUrl: "https://github.com/invoke-ai/InvokeAI",
+    mainTask: "text-to-image",
+    displayOnModelPage: (model) => model.library_name === "diffusers" && model.pipeline_tag === "text-to-image",
+    deeplink: (model) => new URL(`https://models.invoke.ai/huggingface/${model.id}`)
   }
 };