@huggingface/tasks 0.10.22 → 0.11.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (38) hide show
  1. package/dist/index.cjs +96 -41
  2. package/dist/index.js +96 -40
  3. package/dist/scripts/inference-tei-import.d.ts +2 -0
  4. package/dist/scripts/inference-tei-import.d.ts.map +1 -0
  5. package/dist/src/index.d.ts +0 -1
  6. package/dist/src/index.d.ts.map +1 -1
  7. package/dist/src/model-data.d.ts +2 -15
  8. package/dist/src/model-data.d.ts.map +1 -1
  9. package/dist/src/model-libraries-snippets.d.ts +1 -0
  10. package/dist/src/model-libraries-snippets.d.ts.map +1 -1
  11. package/dist/src/model-libraries.d.ts +37 -2
  12. package/dist/src/model-libraries.d.ts.map +1 -1
  13. package/dist/src/tasks/depth-estimation/data.d.ts.map +1 -1
  14. package/dist/src/tasks/feature-extraction/data.d.ts.map +1 -1
  15. package/dist/src/tasks/feature-extraction/inference.d.ts +22 -7
  16. package/dist/src/tasks/feature-extraction/inference.d.ts.map +1 -1
  17. package/dist/src/tasks/object-detection/data.d.ts.map +1 -1
  18. package/dist/src/tasks/zero-shot-image-classification/data.d.ts.map +1 -1
  19. package/dist/src/tasks/zero-shot-object-detection/data.d.ts.map +1 -1
  20. package/package.json +3 -2
  21. package/src/index.ts +0 -1
  22. package/src/model-data.ts +2 -16
  23. package/src/model-libraries-snippets.ts +10 -0
  24. package/src/model-libraries.ts +35 -0
  25. package/src/tasks/depth-estimation/about.md +10 -1
  26. package/src/tasks/depth-estimation/data.ts +13 -9
  27. package/src/tasks/feature-extraction/about.md +46 -1
  28. package/src/tasks/feature-extraction/data.ts +9 -4
  29. package/src/tasks/feature-extraction/inference.ts +23 -5
  30. package/src/tasks/feature-extraction/spec/input.json +34 -13
  31. package/src/tasks/feature-extraction/spec/output.json +10 -2
  32. package/src/tasks/image-text-to-text/data.ts +1 -1
  33. package/src/tasks/object-detection/data.ts +13 -6
  34. package/src/tasks/text-generation/data.ts +1 -1
  35. package/src/tasks/text-to-image/data.ts +4 -4
  36. package/src/tasks/zero-shot-image-classification/about.md +2 -3
  37. package/src/tasks/zero-shot-image-classification/data.ts +4 -0
  38. package/src/tasks/zero-shot-object-detection/data.ts +8 -3
package/dist/index.cjs CHANGED
@@ -24,7 +24,6 @@ __export(src_exports, {
24
24
  ALL_MODEL_LIBRARY_KEYS: () => ALL_MODEL_LIBRARY_KEYS,
25
25
  DATASET_LIBRARIES_UI_ELEMENTS: () => DATASET_LIBRARIES_UI_ELEMENTS,
26
26
  DEFAULT_MEMORY_OPTIONS: () => DEFAULT_MEMORY_OPTIONS,
27
- InferenceDisplayability: () => InferenceDisplayability,
28
27
  LIBRARY_TASK_MAPPING: () => LIBRARY_TASK_MAPPING,
29
28
  LOCAL_APPS: () => LOCAL_APPS,
30
29
  MAPPING_DEFAULT_WIDGET: () => MAPPING_DEFAULT_WIDGET,
@@ -1720,14 +1719,19 @@ var taskData5 = {
1720
1719
  models: [
1721
1720
  {
1722
1721
  description: "A powerful feature extraction model for natural language processing tasks.",
1723
- id: "facebook/bart-base"
1722
+ id: "thenlper/gte-large"
1724
1723
  },
1725
1724
  {
1726
- description: "A strong feature extraction model for coding tasks.",
1727
- id: "microsoft/codebert-base"
1725
+ description: "A strong feature extraction model for retrieval.",
1726
+ id: "Alibaba-NLP/gte-Qwen1.5-7B-instruct"
1727
+ }
1728
+ ],
1729
+ spaces: [
1730
+ {
1731
+ description: "A leaderboard to rank best feature extraction models..",
1732
+ id: "mteb/leaderboard"
1728
1733
  }
1729
1734
  ],
1730
- spaces: [],
1731
1735
  summary: "Feature extraction is the task of extracting features learnt in a model.",
1732
1736
  widgetModels: ["facebook/bart-base"]
1733
1737
  };
@@ -2141,7 +2145,7 @@ var taskData11 = {
2141
2145
  demo: {
2142
2146
  inputs: [
2143
2147
  {
2144
- filename: "mask-generation-input.png",
2148
+ filename: "image-text-to-text-input.png",
2145
2149
  type: "img"
2146
2150
  },
2147
2151
  {
@@ -2356,9 +2360,12 @@ var data_default13 = taskData13;
2356
2360
  var taskData14 = {
2357
2361
  datasets: [
2358
2362
  {
2359
- // TODO write proper description
2360
- description: "Widely used benchmark dataset for multiple Vision tasks.",
2363
+ description: "Widely used benchmark dataset for multiple vision tasks.",
2361
2364
  id: "merve/coco2017"
2365
+ },
2366
+ {
2367
+ description: "Multi-task computer vision benchmark.",
2368
+ id: "merve/pascal-voc"
2362
2369
  }
2363
2370
  ],
2364
2371
  demo: {
@@ -2397,6 +2404,10 @@ var taskData14 = {
2397
2404
  {
2398
2405
  description: "Strong object detection model trained on ImageNet-21k dataset.",
2399
2406
  id: "microsoft/beit-base-patch16-224-pt22k-ft22k"
2407
+ },
2408
+ {
2409
+ description: "Fast and accurate object detection model trained on COCO dataset.",
2410
+ id: "PekingU/rtdetr_r18vd_coco_o365"
2400
2411
  }
2401
2412
  ],
2402
2413
  spaces: [
@@ -2404,10 +2415,6 @@ var taskData14 = {
2404
2415
  description: "Leaderboard to compare various object detection models across several metrics.",
2405
2416
  id: "hf-vision/object_detection_leaderboard"
2406
2417
  },
2407
- {
2408
- description: "An object detection application that can detect unseen objects out of the box.",
2409
- id: "merve/owlv2"
2410
- },
2411
2418
  {
2412
2419
  description: "An application that contains various object detection models to try from.",
2413
2420
  id: "Gradio-Blocks/Object-Detection-With-DETR-and-YOLOS"
@@ -2419,6 +2426,10 @@ var taskData14 = {
2419
2426
  {
2420
2427
  description: "An object tracking, segmentation and inpainting application.",
2421
2428
  id: "VIPLab/Track-Anything"
2429
+ },
2430
+ {
2431
+ description: "Very fast object tracking application based on object detection.",
2432
+ id: "merve/RT-DETR-tracking-coco"
2422
2433
  }
2423
2434
  ],
2424
2435
  summary: "Object Detection models allow users to identify objects of certain defined classes. Object detection models receive an image as input and output the images with bounding boxes and labels on detected objects.",
@@ -2431,8 +2442,12 @@ var data_default14 = taskData14;
2431
2442
  var taskData15 = {
2432
2443
  datasets: [
2433
2444
  {
2434
- description: "NYU Depth V2 Dataset: Video dataset containing both RGB and depth sensor data",
2445
+ description: "NYU Depth V2 Dataset: Video dataset containing both RGB and depth sensor data.",
2435
2446
  id: "sayakpaul/nyu_depth_v2"
2447
+ },
2448
+ {
2449
+ description: "Monocular depth estimation benchmark based without noise and errors.",
2450
+ id: "depth-anything/DA-2K"
2436
2451
  }
2437
2452
  ],
2438
2453
  demo: {
@@ -2452,16 +2467,16 @@ var taskData15 = {
2452
2467
  metrics: [],
2453
2468
  models: [
2454
2469
  {
2455
- description: "Strong Depth Estimation model trained on 1.4 million images.",
2456
- id: "Intel/dpt-large"
2457
- },
2458
- {
2459
- description: "Strong Depth Estimation model trained on a big compilation of datasets.",
2460
- id: "LiheYoung/depth-anything-large-hf"
2470
+ description: "Cutting-edge depth estimation model.",
2471
+ id: "depth-anything/Depth-Anything-V2-Large"
2461
2472
  },
2462
2473
  {
2463
2474
  description: "A strong monocular depth estimation model.",
2464
2475
  id: "Bingxin/Marigold"
2476
+ },
2477
+ {
2478
+ description: "A metric depth estimation model trained on NYU dataset.",
2479
+ id: "Intel/zoedepth-nyu"
2465
2480
  }
2466
2481
  ],
2467
2482
  spaces: [
@@ -2470,8 +2485,8 @@ var taskData15 = {
2470
2485
  id: "radames/dpt-depth-estimation-3d-voxels"
2471
2486
  },
2472
2487
  {
2473
- description: "An application to compare the outputs of different depth estimation models.",
2474
- id: "LiheYoung/Depth-Anything"
2488
+ description: "An application on cutting-edge depth estimation.",
2489
+ id: "depth-anything/Depth-Anything-V2"
2475
2490
  },
2476
2491
  {
2477
2492
  description: "An application to try state-of-the-art depth estimation.",
@@ -3031,18 +3046,18 @@ var taskData24 = {
3031
3046
  id: "latent-consistency/lcm-lora-sdxl"
3032
3047
  },
3033
3048
  {
3034
- description: "A text-to-image model that can generate coherent text inside image.",
3035
- id: "DeepFloyd/IF-I-XL-v1.0"
3049
+ description: "A very fast text-to-image model.",
3050
+ id: "ByteDance/SDXL-Lightning"
3036
3051
  },
3037
3052
  {
3038
3053
  description: "A powerful text-to-image model.",
3039
- id: "kakaobrain/karlo-v1-alpha"
3054
+ id: "stabilityai/stable-diffusion-3-medium-diffusers"
3040
3055
  }
3041
3056
  ],
3042
3057
  spaces: [
3043
3058
  {
3044
3059
  description: "A powerful text-to-image application.",
3045
- id: "stabilityai/stable-diffusion"
3060
+ id: "stabilityai/stable-diffusion-3-medium"
3046
3061
  },
3047
3062
  {
3048
3063
  description: "A text-to-image application to generate comics.",
@@ -3454,7 +3469,7 @@ var taskData29 = {
3454
3469
  spaces: [
3455
3470
  {
3456
3471
  description: "A leaderboard to compare different open-source text generation models based on various benchmarks.",
3457
- id: "HuggingFaceH4/open_llm_leaderboard"
3472
+ id: "open-llm-leaderboard/open_llm_leaderboard"
3458
3473
  },
3459
3474
  {
3460
3475
  description: "An text generation based application based on a very powerful LLaMA2 model.",
@@ -3932,6 +3947,10 @@ var taskData35 = {
3932
3947
  description: "Strong zero-shot image classification model.",
3933
3948
  id: "google/siglip-base-patch16-224"
3934
3949
  },
3950
+ {
3951
+ description: "Small yet powerful zero-shot image classification model that can run on edge devices.",
3952
+ id: "apple/MobileCLIP-S1-OpenCLIP"
3953
+ },
3935
3954
  {
3936
3955
  description: "Strong image classification model for biomedical domain.",
3937
3956
  id: "microsoft/BiomedCLIP-PubMedBERT_256-vit_base_patch16_224"
@@ -3991,11 +4010,11 @@ var taskData36 = {
3991
4010
  ],
3992
4011
  models: [
3993
4012
  {
3994
- description: "Solid zero-shot object detection model that uses CLIP as backbone.",
3995
- id: "google/owlvit-base-patch32"
4013
+ description: "Solid zero-shot object detection model.",
4014
+ id: "IDEA-Research/grounding-dino-base"
3996
4015
  },
3997
4016
  {
3998
- description: "The improved version of the owlvit model.",
4017
+ description: "Cutting-edge zero-shot object detection model.",
3999
4018
  id: "google/owlv2-base-patch16-ensemble"
4000
4019
  }
4001
4020
  ],
@@ -4003,6 +4022,10 @@ var taskData36 = {
4003
4022
  {
4004
4023
  description: "A demo to try the state-of-the-art zero-shot object detection model, OWLv2.",
4005
4024
  id: "merve/owlv2"
4025
+ },
4026
+ {
4027
+ description: "A demo that combines a zero-shot object detection and mask generation model for zero-shot segmentation.",
4028
+ id: "merve/OWLSAM"
4006
4029
  }
4007
4030
  ],
4008
4031
  summary: "Zero-shot object detection is a computer vision task to detect objects and their classes in images, without any prior training or knowledge of the classes. Zero-shot object detection models receive an image as input, as well as a list of candidate classes, and output the bounding boxes and labels where the objects have been detected.",
@@ -4330,6 +4353,15 @@ var mars5_tts = (model) => [
4330
4353
  from inference import Mars5TTS
4331
4354
  mars5 = Mars5TTS.from_pretrained("${model.id}")`
4332
4355
  ];
4356
+ var mesh_anything = () => [
4357
+ `# Install from https://github.com/buaacyw/MeshAnything.git
4358
+
4359
+ from MeshAnything.models.meshanything import MeshAnything
4360
+
4361
+ # refer to https://github.com/buaacyw/MeshAnything/blob/main/main.py#L91 on how to define args
4362
+ # and https://github.com/buaacyw/MeshAnything/blob/main/app.py regarding usage
4363
+ model = MeshAnything(args)`
4364
+ ];
4333
4365
  var open_clip = (model) => [
4334
4366
  `import open_clip
4335
4367
 
@@ -4858,6 +4890,12 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4858
4890
  filter: false,
4859
4891
  countDownloads: `path:"params.index.json"`
4860
4892
  },
4893
+ champ: {
4894
+ prettyLabel: "Champ",
4895
+ repoName: "Champ",
4896
+ repoUrl: "https://github.com/fudan-generative-vision/champ",
4897
+ countDownloads: `path:"champ/motion_module.pth"`
4898
+ },
4861
4899
  chat_tts: {
4862
4900
  prettyLabel: "ChatTTS",
4863
4901
  repoName: "ChatTTS",
@@ -4866,6 +4904,13 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4866
4904
  filter: false,
4867
4905
  countDownloads: `path:"asset/GPT.pt"`
4868
4906
  },
4907
+ colpali: {
4908
+ prettyLabel: "ColPali",
4909
+ repoName: "ColPali",
4910
+ repoUrl: "https://github.com/ManuelFay/colpali",
4911
+ filter: false,
4912
+ countDownloads: `path:"adapter_config.json"`
4913
+ },
4869
4914
  diffusers: {
4870
4915
  prettyLabel: "Diffusers",
4871
4916
  repoName: "\u{1F917}/diffusers",
@@ -4965,6 +5010,12 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4965
5010
  filter: false,
4966
5011
  countDownloads: `path:"ckpt/tensor00000_000" OR path:"ckpt-0/tensor00000_000"`
4967
5012
  },
5013
+ hallo: {
5014
+ prettyLabel: "Hallo",
5015
+ repoName: "Hallo",
5016
+ repoUrl: "https://github.com/fudan-generative-vision/hallo",
5017
+ countDownloads: `path:"hallo/net.pth"`
5018
+ },
4968
5019
  "hunyuan-dit": {
4969
5020
  prettyLabel: "HunyuanDiT",
4970
5021
  repoName: "HunyuanDiT",
@@ -5015,6 +5066,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
5015
5066
  countDownloads: `path:"mars5_ar.safetensors"`,
5016
5067
  snippets: mars5_tts
5017
5068
  },
5069
+ "mesh-anything": {
5070
+ prettyLabel: "MeshAnything",
5071
+ repoName: "MeshAnything",
5072
+ repoUrl: "https://github.com/buaacyw/MeshAnything",
5073
+ filter: false,
5074
+ countDownloads: `path:"MeshAnything_350m.pth"`,
5075
+ snippets: mesh_anything
5076
+ },
5018
5077
  "ml-agents": {
5019
5078
  prettyLabel: "ml-agents",
5020
5079
  repoName: "ml-agents",
@@ -5040,6 +5099,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
5040
5099
  filter: false,
5041
5100
  countDownloads: `path:"model.safetensors"`
5042
5101
  },
5102
+ "mlc-llm": {
5103
+ prettyLabel: "MLC-LLM",
5104
+ repoName: "MLC-LLM",
5105
+ repoUrl: "https://github.com/mlc-ai/mlc-llm",
5106
+ docsUrl: "https://llm.mlc.ai/docs/",
5107
+ filter: false,
5108
+ countDownloads: `path:"mlc-chat-config.json"`
5109
+ },
5043
5110
  nemo: {
5044
5111
  prettyLabel: "NeMo",
5045
5112
  repoName: "NeMo",
@@ -5258,17 +5325,6 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
5258
5325
  var ALL_MODEL_LIBRARY_KEYS = Object.keys(MODEL_LIBRARIES_UI_ELEMENTS);
5259
5326
  var ALL_DISPLAY_MODEL_LIBRARY_KEYS = Object.entries(MODEL_LIBRARIES_UI_ELEMENTS).filter(([_, v]) => v.filter).map(([k]) => k);
5260
5327
 
5261
- // src/model-data.ts
5262
- var InferenceDisplayability = /* @__PURE__ */ ((InferenceDisplayability2) => {
5263
- InferenceDisplayability2["Yes"] = "Yes";
5264
- InferenceDisplayability2["ExplicitOptOut"] = "ExplicitOptOut";
5265
- InferenceDisplayability2["CustomCode"] = "CustomCode";
5266
- InferenceDisplayability2["LibraryNotDetected"] = "LibraryNotDetected";
5267
- InferenceDisplayability2["PipelineNotDetected"] = "PipelineNotDetected";
5268
- InferenceDisplayability2["PipelineLibraryPairNotSupported"] = "PipelineLibraryPairNotSupported";
5269
- return InferenceDisplayability2;
5270
- })(InferenceDisplayability || {});
5271
-
5272
5328
  // src/tokenizer-data.ts
5273
5329
  var SPECIAL_TOKENS_ATTRIBUTES = [
5274
5330
  "bos_token",
@@ -6361,7 +6417,6 @@ var DATASET_LIBRARIES_UI_ELEMENTS = {
6361
6417
  ALL_MODEL_LIBRARY_KEYS,
6362
6418
  DATASET_LIBRARIES_UI_ELEMENTS,
6363
6419
  DEFAULT_MEMORY_OPTIONS,
6364
- InferenceDisplayability,
6365
6420
  LIBRARY_TASK_MAPPING,
6366
6421
  LOCAL_APPS,
6367
6422
  MAPPING_DEFAULT_WIDGET,
package/dist/index.js CHANGED
@@ -1681,14 +1681,19 @@ var taskData5 = {
1681
1681
  models: [
1682
1682
  {
1683
1683
  description: "A powerful feature extraction model for natural language processing tasks.",
1684
- id: "facebook/bart-base"
1684
+ id: "thenlper/gte-large"
1685
1685
  },
1686
1686
  {
1687
- description: "A strong feature extraction model for coding tasks.",
1688
- id: "microsoft/codebert-base"
1687
+ description: "A strong feature extraction model for retrieval.",
1688
+ id: "Alibaba-NLP/gte-Qwen1.5-7B-instruct"
1689
+ }
1690
+ ],
1691
+ spaces: [
1692
+ {
1693
+ description: "A leaderboard to rank best feature extraction models..",
1694
+ id: "mteb/leaderboard"
1689
1695
  }
1690
1696
  ],
1691
- spaces: [],
1692
1697
  summary: "Feature extraction is the task of extracting features learnt in a model.",
1693
1698
  widgetModels: ["facebook/bart-base"]
1694
1699
  };
@@ -2102,7 +2107,7 @@ var taskData11 = {
2102
2107
  demo: {
2103
2108
  inputs: [
2104
2109
  {
2105
- filename: "mask-generation-input.png",
2110
+ filename: "image-text-to-text-input.png",
2106
2111
  type: "img"
2107
2112
  },
2108
2113
  {
@@ -2317,9 +2322,12 @@ var data_default13 = taskData13;
2317
2322
  var taskData14 = {
2318
2323
  datasets: [
2319
2324
  {
2320
- // TODO write proper description
2321
- description: "Widely used benchmark dataset for multiple Vision tasks.",
2325
+ description: "Widely used benchmark dataset for multiple vision tasks.",
2322
2326
  id: "merve/coco2017"
2327
+ },
2328
+ {
2329
+ description: "Multi-task computer vision benchmark.",
2330
+ id: "merve/pascal-voc"
2323
2331
  }
2324
2332
  ],
2325
2333
  demo: {
@@ -2358,6 +2366,10 @@ var taskData14 = {
2358
2366
  {
2359
2367
  description: "Strong object detection model trained on ImageNet-21k dataset.",
2360
2368
  id: "microsoft/beit-base-patch16-224-pt22k-ft22k"
2369
+ },
2370
+ {
2371
+ description: "Fast and accurate object detection model trained on COCO dataset.",
2372
+ id: "PekingU/rtdetr_r18vd_coco_o365"
2361
2373
  }
2362
2374
  ],
2363
2375
  spaces: [
@@ -2365,10 +2377,6 @@ var taskData14 = {
2365
2377
  description: "Leaderboard to compare various object detection models across several metrics.",
2366
2378
  id: "hf-vision/object_detection_leaderboard"
2367
2379
  },
2368
- {
2369
- description: "An object detection application that can detect unseen objects out of the box.",
2370
- id: "merve/owlv2"
2371
- },
2372
2380
  {
2373
2381
  description: "An application that contains various object detection models to try from.",
2374
2382
  id: "Gradio-Blocks/Object-Detection-With-DETR-and-YOLOS"
@@ -2380,6 +2388,10 @@ var taskData14 = {
2380
2388
  {
2381
2389
  description: "An object tracking, segmentation and inpainting application.",
2382
2390
  id: "VIPLab/Track-Anything"
2391
+ },
2392
+ {
2393
+ description: "Very fast object tracking application based on object detection.",
2394
+ id: "merve/RT-DETR-tracking-coco"
2383
2395
  }
2384
2396
  ],
2385
2397
  summary: "Object Detection models allow users to identify objects of certain defined classes. Object detection models receive an image as input and output the images with bounding boxes and labels on detected objects.",
@@ -2392,8 +2404,12 @@ var data_default14 = taskData14;
2392
2404
  var taskData15 = {
2393
2405
  datasets: [
2394
2406
  {
2395
- description: "NYU Depth V2 Dataset: Video dataset containing both RGB and depth sensor data",
2407
+ description: "NYU Depth V2 Dataset: Video dataset containing both RGB and depth sensor data.",
2396
2408
  id: "sayakpaul/nyu_depth_v2"
2409
+ },
2410
+ {
2411
+ description: "Monocular depth estimation benchmark based without noise and errors.",
2412
+ id: "depth-anything/DA-2K"
2397
2413
  }
2398
2414
  ],
2399
2415
  demo: {
@@ -2413,16 +2429,16 @@ var taskData15 = {
2413
2429
  metrics: [],
2414
2430
  models: [
2415
2431
  {
2416
- description: "Strong Depth Estimation model trained on 1.4 million images.",
2417
- id: "Intel/dpt-large"
2418
- },
2419
- {
2420
- description: "Strong Depth Estimation model trained on a big compilation of datasets.",
2421
- id: "LiheYoung/depth-anything-large-hf"
2432
+ description: "Cutting-edge depth estimation model.",
2433
+ id: "depth-anything/Depth-Anything-V2-Large"
2422
2434
  },
2423
2435
  {
2424
2436
  description: "A strong monocular depth estimation model.",
2425
2437
  id: "Bingxin/Marigold"
2438
+ },
2439
+ {
2440
+ description: "A metric depth estimation model trained on NYU dataset.",
2441
+ id: "Intel/zoedepth-nyu"
2426
2442
  }
2427
2443
  ],
2428
2444
  spaces: [
@@ -2431,8 +2447,8 @@ var taskData15 = {
2431
2447
  id: "radames/dpt-depth-estimation-3d-voxels"
2432
2448
  },
2433
2449
  {
2434
- description: "An application to compare the outputs of different depth estimation models.",
2435
- id: "LiheYoung/Depth-Anything"
2450
+ description: "An application on cutting-edge depth estimation.",
2451
+ id: "depth-anything/Depth-Anything-V2"
2436
2452
  },
2437
2453
  {
2438
2454
  description: "An application to try state-of-the-art depth estimation.",
@@ -2992,18 +3008,18 @@ var taskData24 = {
2992
3008
  id: "latent-consistency/lcm-lora-sdxl"
2993
3009
  },
2994
3010
  {
2995
- description: "A text-to-image model that can generate coherent text inside image.",
2996
- id: "DeepFloyd/IF-I-XL-v1.0"
3011
+ description: "A very fast text-to-image model.",
3012
+ id: "ByteDance/SDXL-Lightning"
2997
3013
  },
2998
3014
  {
2999
3015
  description: "A powerful text-to-image model.",
3000
- id: "kakaobrain/karlo-v1-alpha"
3016
+ id: "stabilityai/stable-diffusion-3-medium-diffusers"
3001
3017
  }
3002
3018
  ],
3003
3019
  spaces: [
3004
3020
  {
3005
3021
  description: "A powerful text-to-image application.",
3006
- id: "stabilityai/stable-diffusion"
3022
+ id: "stabilityai/stable-diffusion-3-medium"
3007
3023
  },
3008
3024
  {
3009
3025
  description: "A text-to-image application to generate comics.",
@@ -3415,7 +3431,7 @@ var taskData29 = {
3415
3431
  spaces: [
3416
3432
  {
3417
3433
  description: "A leaderboard to compare different open-source text generation models based on various benchmarks.",
3418
- id: "HuggingFaceH4/open_llm_leaderboard"
3434
+ id: "open-llm-leaderboard/open_llm_leaderboard"
3419
3435
  },
3420
3436
  {
3421
3437
  description: "An text generation based application based on a very powerful LLaMA2 model.",
@@ -3893,6 +3909,10 @@ var taskData35 = {
3893
3909
  description: "Strong zero-shot image classification model.",
3894
3910
  id: "google/siglip-base-patch16-224"
3895
3911
  },
3912
+ {
3913
+ description: "Small yet powerful zero-shot image classification model that can run on edge devices.",
3914
+ id: "apple/MobileCLIP-S1-OpenCLIP"
3915
+ },
3896
3916
  {
3897
3917
  description: "Strong image classification model for biomedical domain.",
3898
3918
  id: "microsoft/BiomedCLIP-PubMedBERT_256-vit_base_patch16_224"
@@ -3952,11 +3972,11 @@ var taskData36 = {
3952
3972
  ],
3953
3973
  models: [
3954
3974
  {
3955
- description: "Solid zero-shot object detection model that uses CLIP as backbone.",
3956
- id: "google/owlvit-base-patch32"
3975
+ description: "Solid zero-shot object detection model.",
3976
+ id: "IDEA-Research/grounding-dino-base"
3957
3977
  },
3958
3978
  {
3959
- description: "The improved version of the owlvit model.",
3979
+ description: "Cutting-edge zero-shot object detection model.",
3960
3980
  id: "google/owlv2-base-patch16-ensemble"
3961
3981
  }
3962
3982
  ],
@@ -3964,6 +3984,10 @@ var taskData36 = {
3964
3984
  {
3965
3985
  description: "A demo to try the state-of-the-art zero-shot object detection model, OWLv2.",
3966
3986
  id: "merve/owlv2"
3987
+ },
3988
+ {
3989
+ description: "A demo that combines a zero-shot object detection and mask generation model for zero-shot segmentation.",
3990
+ id: "merve/OWLSAM"
3967
3991
  }
3968
3992
  ],
3969
3993
  summary: "Zero-shot object detection is a computer vision task to detect objects and their classes in images, without any prior training or knowledge of the classes. Zero-shot object detection models receive an image as input, as well as a list of candidate classes, and output the bounding boxes and labels where the objects have been detected.",
@@ -4291,6 +4315,15 @@ var mars5_tts = (model) => [
4291
4315
  from inference import Mars5TTS
4292
4316
  mars5 = Mars5TTS.from_pretrained("${model.id}")`
4293
4317
  ];
4318
+ var mesh_anything = () => [
4319
+ `# Install from https://github.com/buaacyw/MeshAnything.git
4320
+
4321
+ from MeshAnything.models.meshanything import MeshAnything
4322
+
4323
+ # refer to https://github.com/buaacyw/MeshAnything/blob/main/main.py#L91 on how to define args
4324
+ # and https://github.com/buaacyw/MeshAnything/blob/main/app.py regarding usage
4325
+ model = MeshAnything(args)`
4326
+ ];
4294
4327
  var open_clip = (model) => [
4295
4328
  `import open_clip
4296
4329
 
@@ -4819,6 +4852,12 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4819
4852
  filter: false,
4820
4853
  countDownloads: `path:"params.index.json"`
4821
4854
  },
4855
+ champ: {
4856
+ prettyLabel: "Champ",
4857
+ repoName: "Champ",
4858
+ repoUrl: "https://github.com/fudan-generative-vision/champ",
4859
+ countDownloads: `path:"champ/motion_module.pth"`
4860
+ },
4822
4861
  chat_tts: {
4823
4862
  prettyLabel: "ChatTTS",
4824
4863
  repoName: "ChatTTS",
@@ -4827,6 +4866,13 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4827
4866
  filter: false,
4828
4867
  countDownloads: `path:"asset/GPT.pt"`
4829
4868
  },
4869
+ colpali: {
4870
+ prettyLabel: "ColPali",
4871
+ repoName: "ColPali",
4872
+ repoUrl: "https://github.com/ManuelFay/colpali",
4873
+ filter: false,
4874
+ countDownloads: `path:"adapter_config.json"`
4875
+ },
4830
4876
  diffusers: {
4831
4877
  prettyLabel: "Diffusers",
4832
4878
  repoName: "\u{1F917}/diffusers",
@@ -4926,6 +4972,12 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4926
4972
  filter: false,
4927
4973
  countDownloads: `path:"ckpt/tensor00000_000" OR path:"ckpt-0/tensor00000_000"`
4928
4974
  },
4975
+ hallo: {
4976
+ prettyLabel: "Hallo",
4977
+ repoName: "Hallo",
4978
+ repoUrl: "https://github.com/fudan-generative-vision/hallo",
4979
+ countDownloads: `path:"hallo/net.pth"`
4980
+ },
4929
4981
  "hunyuan-dit": {
4930
4982
  prettyLabel: "HunyuanDiT",
4931
4983
  repoName: "HunyuanDiT",
@@ -4976,6 +5028,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
4976
5028
  countDownloads: `path:"mars5_ar.safetensors"`,
4977
5029
  snippets: mars5_tts
4978
5030
  },
5031
+ "mesh-anything": {
5032
+ prettyLabel: "MeshAnything",
5033
+ repoName: "MeshAnything",
5034
+ repoUrl: "https://github.com/buaacyw/MeshAnything",
5035
+ filter: false,
5036
+ countDownloads: `path:"MeshAnything_350m.pth"`,
5037
+ snippets: mesh_anything
5038
+ },
4979
5039
  "ml-agents": {
4980
5040
  prettyLabel: "ml-agents",
4981
5041
  repoName: "ml-agents",
@@ -5001,6 +5061,14 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
5001
5061
  filter: false,
5002
5062
  countDownloads: `path:"model.safetensors"`
5003
5063
  },
5064
+ "mlc-llm": {
5065
+ prettyLabel: "MLC-LLM",
5066
+ repoName: "MLC-LLM",
5067
+ repoUrl: "https://github.com/mlc-ai/mlc-llm",
5068
+ docsUrl: "https://llm.mlc.ai/docs/",
5069
+ filter: false,
5070
+ countDownloads: `path:"mlc-chat-config.json"`
5071
+ },
5004
5072
  nemo: {
5005
5073
  prettyLabel: "NeMo",
5006
5074
  repoName: "NeMo",
@@ -5219,17 +5287,6 @@ var MODEL_LIBRARIES_UI_ELEMENTS = {
5219
5287
  var ALL_MODEL_LIBRARY_KEYS = Object.keys(MODEL_LIBRARIES_UI_ELEMENTS);
5220
5288
  var ALL_DISPLAY_MODEL_LIBRARY_KEYS = Object.entries(MODEL_LIBRARIES_UI_ELEMENTS).filter(([_, v]) => v.filter).map(([k]) => k);
5221
5289
 
5222
- // src/model-data.ts
5223
- var InferenceDisplayability = /* @__PURE__ */ ((InferenceDisplayability2) => {
5224
- InferenceDisplayability2["Yes"] = "Yes";
5225
- InferenceDisplayability2["ExplicitOptOut"] = "ExplicitOptOut";
5226
- InferenceDisplayability2["CustomCode"] = "CustomCode";
5227
- InferenceDisplayability2["LibraryNotDetected"] = "LibraryNotDetected";
5228
- InferenceDisplayability2["PipelineNotDetected"] = "PipelineNotDetected";
5229
- InferenceDisplayability2["PipelineLibraryPairNotSupported"] = "PipelineLibraryPairNotSupported";
5230
- return InferenceDisplayability2;
5231
- })(InferenceDisplayability || {});
5232
-
5233
5290
  // src/tokenizer-data.ts
5234
5291
  var SPECIAL_TOKENS_ATTRIBUTES = [
5235
5292
  "bos_token",
@@ -6321,7 +6378,6 @@ export {
6321
6378
  ALL_MODEL_LIBRARY_KEYS,
6322
6379
  DATASET_LIBRARIES_UI_ELEMENTS,
6323
6380
  DEFAULT_MEMORY_OPTIONS,
6324
- InferenceDisplayability,
6325
6381
  LIBRARY_TASK_MAPPING,
6326
6382
  LOCAL_APPS,
6327
6383
  MAPPING_DEFAULT_WIDGET,
@@ -0,0 +1,2 @@
1
+ export {};
2
+ //# sourceMappingURL=inference-tei-import.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"inference-tei-import.d.ts","sourceRoot":"","sources":["../../scripts/inference-tei-import.ts"],"names":[],"mappings":""}
@@ -8,7 +8,6 @@ export type { LibraryUiElement, ModelLibraryKey } from "./model-libraries";
8
8
  export type { ModelData, TransformersInfo } from "./model-data";
9
9
  export type { AddedToken, SpecialTokensMap, TokenizerConfig } from "./tokenizer-data";
10
10
  export type { WidgetExample, WidgetExampleAttribute, WidgetExampleAssetAndPromptInput, WidgetExampleAssetAndTextInput, WidgetExampleAssetAndZeroShotInput, WidgetExampleAssetInput, WidgetExampleChatInput, WidgetExampleSentenceSimilarityInput, WidgetExampleStructuredDataInput, WidgetExampleTableDataInput, WidgetExampleTextAndContextInput, WidgetExampleTextAndTableInput, WidgetExampleTextInput, WidgetExampleZeroShotTextInput, WidgetExampleOutput, WidgetExampleOutputUrl, WidgetExampleOutputLabels, WidgetExampleOutputAnswerScore, WidgetExampleOutputText, } from "./widget-example";
11
- export { InferenceDisplayability } from "./model-data";
12
11
  export { SPECIAL_TOKENS_ATTRIBUTES } from "./tokenizer-data";
13
12
  import * as snippets from "./snippets";
14
13
  export { snippets };
@@ -1 +1 @@
1
- {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,oBAAoB,EAAE,MAAM,oBAAoB,CAAC;AAC1D,OAAO,EAAE,sBAAsB,EAAE,MAAM,yBAAyB,CAAC;AACjE,YAAY,EAAE,QAAQ,EAAE,QAAQ,EAAE,aAAa,EAAE,WAAW,EAAE,MAAM,SAAS,CAAC;AAC9E,cAAc,SAAS,CAAC;AACxB,OAAO,EACN,aAAa,EACb,cAAc,EACd,KAAK,UAAU,EACf,KAAK,YAAY,EACjB,KAAK,YAAY,EACjB,KAAK,QAAQ,EACb,UAAU,EACV,eAAe,EACf,aAAa,EACb,kBAAkB,GAClB,MAAM,aAAa,CAAC;AACrB,OAAO,EAAE,8BAA8B,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,MAAM,mBAAmB,CAAC;AACxH,YAAY,EAAE,gBAAgB,EAAE,eAAe,EAAE,MAAM,mBAAmB,CAAC;AAC3E,YAAY,EAAE,SAAS,EAAE,gBAAgB,EAAE,MAAM,cAAc,CAAC;AAChE,YAAY,EAAE,UAAU,EAAE,gBAAgB,EAAE,eAAe,EAAE,MAAM,kBAAkB,CAAC;AACtF,YAAY,EACX,aAAa,EACb,sBAAsB,EACtB,gCAAgC,EAChC,8BAA8B,EAC9B,kCAAkC,EAClC,uBAAuB,EACvB,sBAAsB,EACtB,oCAAoC,EACpC,gCAAgC,EAChC,2BAA2B,EAC3B,gCAAgC,EAChC,8BAA8B,EAC9B,sBAAsB,EACtB,8BAA8B,EAC9B,mBAAmB,EACnB,sBAAsB,EACtB,yBAAyB,EACzB,8BAA8B,EAC9B,uBAAuB,GACvB,MAAM,kBAAkB,CAAC;AAC1B,OAAO,EAAE,uBAAuB,EAAE,MAAM,cAAc,CAAC;AACvD,OAAO,EAAE,yBAAyB,EAAE,MAAM,kBAAkB,CAAC;AAE7D,OAAO,KAAK,QAAQ,MAAM,YAAY,CAAC;AACvC,OAAO,EAAE,QAAQ,EAAE,CAAC;AAEpB,OAAO,EAAE,IAAI,EAAE,sBAAsB,EAAE,MAAM,YAAY,CAAC;AAC1D,YAAY,EAAE,YAAY,EAAE,OAAO,EAAE,MAAM,YAAY,CAAC;AACxD,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAC1C,YAAY,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,cAAc,CAAC;AAE1D,OAAO,EAAE,6BAA6B,EAAE,MAAM,qBAAqB,CAAC;AACpE,YAAY,EAAE,uBAAuB,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC"}
1
+ {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../src/index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,oBAAoB,EAAE,MAAM,oBAAoB,CAAC;AAC1D,OAAO,EAAE,sBAAsB,EAAE,MAAM,yBAAyB,CAAC;AACjE,YAAY,EAAE,QAAQ,EAAE,QAAQ,EAAE,aAAa,EAAE,WAAW,EAAE,MAAM,SAAS,CAAC;AAC9E,cAAc,SAAS,CAAC;AACxB,OAAO,EACN,aAAa,EACb,cAAc,EACd,KAAK,UAAU,EACf,KAAK,YAAY,EACjB,KAAK,YAAY,EACjB,KAAK,QAAQ,EACb,UAAU,EACV,eAAe,EACf,aAAa,EACb,kBAAkB,GAClB,MAAM,aAAa,CAAC;AACrB,OAAO,EAAE,8BAA8B,EAAE,sBAAsB,EAAE,2BAA2B,EAAE,MAAM,mBAAmB,CAAC;AACxH,YAAY,EAAE,gBAAgB,EAAE,eAAe,EAAE,MAAM,mBAAmB,CAAC;AAC3E,YAAY,EAAE,SAAS,EAAE,gBAAgB,EAAE,MAAM,cAAc,CAAC;AAChE,YAAY,EAAE,UAAU,EAAE,gBAAgB,EAAE,eAAe,EAAE,MAAM,kBAAkB,CAAC;AACtF,YAAY,EACX,aAAa,EACb,sBAAsB,EACtB,gCAAgC,EAChC,8BAA8B,EAC9B,kCAAkC,EAClC,uBAAuB,EACvB,sBAAsB,EACtB,oCAAoC,EACpC,gCAAgC,EAChC,2BAA2B,EAC3B,gCAAgC,EAChC,8BAA8B,EAC9B,sBAAsB,EACtB,8BAA8B,EAC9B,mBAAmB,EACnB,sBAAsB,EACtB,yBAAyB,EACzB,8BAA8B,EAC9B,uBAAuB,GACvB,MAAM,kBAAkB,CAAC;AAC1B,OAAO,EAAE,yBAAyB,EAAE,MAAM,kBAAkB,CAAC;AAE7D,OAAO,KAAK,QAAQ,MAAM,YAAY,CAAC;AACvC,OAAO,EAAE,QAAQ,EAAE,CAAC;AAEpB,OAAO,EAAE,IAAI,EAAE,sBAAsB,EAAE,MAAM,YAAY,CAAC;AAC1D,YAAY,EAAE,YAAY,EAAE,OAAO,EAAE,MAAM,YAAY,CAAC;AACxD,OAAO,EAAE,UAAU,EAAE,MAAM,cAAc,CAAC;AAC1C,YAAY,EAAE,QAAQ,EAAE,WAAW,EAAE,MAAM,cAAc,CAAC;AAE1D,OAAO,EAAE,6BAA6B,EAAE,MAAM,qBAAqB,CAAC;AACpE,YAAY,EAAE,uBAAuB,EAAE,iBAAiB,EAAE,MAAM,qBAAqB,CAAC"}
@@ -1,20 +1,6 @@
1
1
  import type { PipelineType } from "./pipelines";
2
2
  import type { WidgetExample } from "./widget-example";
3
3
  import type { TokenizerConfig } from "./tokenizer-data";
4
- export declare enum InferenceDisplayability {
5
- /**
6
- * Yes
7
- */
8
- Yes = "Yes",
9
- /**
10
- * And then, all the possible reasons why it's no:
11
- */
12
- ExplicitOptOut = "ExplicitOptOut",
13
- CustomCode = "CustomCode",
14
- LibraryNotDetected = "LibraryNotDetected",
15
- PipelineNotDetected = "PipelineNotDetected",
16
- PipelineLibraryPairNotSupported = "PipelineLibraryPairNotSupported"
17
- }
18
4
  /**
19
5
  * Public interface for model metadata
20
6
  */
@@ -25,8 +11,9 @@ export interface ModelData {
25
11
  id: string;
26
12
  /**
27
13
  * Whether or not to enable inference widget for this model
14
+ * TODO(type it)
28
15
  */
29
- inference: InferenceDisplayability;
16
+ inference: string;
30
17
  /**
31
18
  * is this model private?
32
19
  */