@huggingface/tasks 0.18.7 → 0.18.8

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.
@@ -430,6 +430,18 @@ export declare const SKUS: {
430
430
  tflops: number;
431
431
  memory: number[];
432
432
  };
433
+ "RX 6650 XT": {
434
+ tflops: number;
435
+ memory: number[];
436
+ };
437
+ "RX 6600 XT": {
438
+ tflops: number;
439
+ memory: number[];
440
+ };
441
+ "RX 6600": {
442
+ tflops: number;
443
+ memory: number[];
444
+ };
433
445
  "Radeon Pro VII": {
434
446
  tflops: number;
435
447
  memory: number[];
@@ -1 +1 @@
1
- {"version":3,"file":"hardware.d.ts","sourceRoot":"","sources":["../../src/hardware.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,eAAO,MAAM,iDAAiD,QAAW,CAAC;AAC1E,eAAO,MAAM,yDAAyD,QAAW,CAAC;AAClF,eAAO,MAAM,oCAAoC,QAAU,CAAC;AAE5D;;;GAGG;AACH,eAAO,MAAM,+CAA+C,QAAW,CAAC;AAExE,MAAM,WAAW,YAAY;IAC5B;;;;;;;;;OASG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,EAAE,CAAC;CAClB;AAED,eAAO,MAAM,sBAAsB,UAAqD,CAAC;AAEzF,eAAO,MAAM,IAAI;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA4mBuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}
1
+ {"version":3,"file":"hardware.d.ts","sourceRoot":"","sources":["../../src/hardware.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,eAAO,MAAM,iDAAiD,QAAW,CAAC;AAC1E,eAAO,MAAM,yDAAyD,QAAW,CAAC;AAClF,eAAO,MAAM,oCAAoC,QAAU,CAAC;AAE5D;;;GAGG;AACH,eAAO,MAAM,+CAA+C,QAAW,CAAC;AAExE,MAAM,WAAW,YAAY;IAC5B;;;;;;;;;OASG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,EAAE,CAAC;CAClB;AAED,eAAO,MAAM,sBAAsB,UAAqD,CAAC;AAEzF,eAAO,MAAM,IAAI;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAwnBuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}
@@ -415,6 +415,18 @@ exports.SKUS = {
415
415
  tflops: 22.58,
416
416
  memory: [10],
417
417
  },
418
+ "RX 6650 XT": {
419
+ tflops: 21.59,
420
+ memory: [8],
421
+ },
422
+ "RX 6600 XT": {
423
+ tflops: 21.21,
424
+ memory: [8],
425
+ },
426
+ "RX 6600": {
427
+ tflops: 17.86,
428
+ memory: [8],
429
+ },
418
430
  "Radeon Pro VII": {
419
431
  tflops: 26.11,
420
432
  memory: [16],
@@ -0,0 +1,4 @@
1
+ import type { TaskDataCustom } from "../index.js";
2
+ declare const taskData: TaskDataCustom;
3
+ export default taskData;
4
+ //# sourceMappingURL=data.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"data.d.ts","sourceRoot":"","sources":["../../../../src/tasks/any-to-any/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,aAAa,CAAC;AAElD,QAAA,MAAM,QAAQ,EAAE,cA6Df,CAAC;AAEF,eAAe,QAAQ,CAAC"}
@@ -0,0 +1,63 @@
1
+ "use strict";
2
+ Object.defineProperty(exports, "__esModule", { value: true });
3
+ const taskData = {
4
+ datasets: [
5
+ {
6
+ description: "A dataset with multiple modality input and output pairs.",
7
+ id: "PKU-Alignment/align-anything",
8
+ },
9
+ ],
10
+ demo: {
11
+ inputs: [
12
+ {
13
+ filename: "any-to-any-input.jpg",
14
+ type: "img",
15
+ },
16
+ {
17
+ label: "Text Prompt",
18
+ content: "Describe the position of the bee in detail.",
19
+ type: "text",
20
+ },
21
+ ],
22
+ outputs: [
23
+ {
24
+ label: "Generated Text",
25
+ content: "The place in the picture is Osaka Castle, located in Osaka, Japan. Osaka Castle is a historic castle that was originally built in the 16th century by Toyotomi Hideyoshi, a powerful warlord of the time. It is one of the most famous landmarks in Osaka and is known for its distinctive white walls and black roof tiles. The castle has been rebuilt several times over the centuries and is now a popular tourist attraction, offering visitors a glimpse into Japan's rich history and culture.",
26
+ type: "text",
27
+ },
28
+ {
29
+ filename: "any-to-any-output.wav",
30
+ type: "audio",
31
+ },
32
+ ],
33
+ },
34
+ metrics: [],
35
+ models: [
36
+ {
37
+ description: "Strong model that can take in video, audio, image, text and output text and natural speech.",
38
+ id: "Qwen/Qwen2.5-Omni-7B",
39
+ },
40
+ {
41
+ description: "Robust model that can take in image and text and generate image and text.",
42
+ id: "deepseek-ai/Janus-Pro-7B",
43
+ },
44
+ {
45
+ description: "Any-to-any model with speech, video, audio, image and text understanding capabilities.",
46
+ id: "openbmb/MiniCPM-o-2_6",
47
+ },
48
+ {
49
+ description: "A model that can understand image and text and generate image and text.",
50
+ id: "EPFL-VILAB/4M-21_XL",
51
+ },
52
+ ],
53
+ spaces: [
54
+ {
55
+ description: "An application to chat with an any-to-any (image & text) model.",
56
+ id: "deepseek-ai/Janus-Pro-7B",
57
+ },
58
+ ],
59
+ summary: "Any-to-any models can understand two or more modalities and output two or more modalities.",
60
+ widgetModels: [],
61
+ youtubeId: "",
62
+ };
63
+ exports.default = taskData;
@@ -1 +1 @@
1
- {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/tasks/index.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAC;AA6CpD,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,6CAA6C,CAAC;AACjE,YAAY,EACX,mBAAmB,EACnB,0BAA0B,EAC1B,oBAAoB,EACpB,4BAA4B,EAC5B,2BAA2B,EAC3B,0BAA0B,EAC1B,gCAAgC,EAChC,+BAA+B,GAC/B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,4CAA4C,CAAC;AAChE,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,0BAA0B,CAAC;AAC9C,YAAY,EACX,wBAAwB,EACxB,yBAAyB,EACzB,gCAAgC,EAChC,6BAA6B,GAC7B,MAAM,qCAAqC,CAAC;AAC7C,mBAAmB,+BAA+B,CAAC;AACnD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,oCAAoC,CAAC;AACxD,mBAAmB,8BAA8B,CAAC;AAClD,mBAAmB,yCAAyC,CAAC;AAC7D,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,sBAAsB,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,+BAA+B,CAAC;AACnH,mBAAmB,qCAAqC,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAC;AACtF,YAAY,EACX,6BAA6B,EAC7B,uBAAuB,EACvB,wBAAwB,EACxB,+BAA+B,EAC/B,4BAA4B,GAC5B,MAAM,oCAAoC,CAAC;AAC5C,YAAY,EACX,gCAAgC,EAChC,gCAAgC,EAChC,mBAAmB,EACnB,oBAAoB,EACpB,2BAA2B,EAC3B,qCAAqC,EACrC,kCAAkC,EAClC,yBAAyB,EACzB,uCAAuC,EACvC,0BAA0B,GAC1B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,0CAA0C,CAAC;AAC9D,mBAAmB,yCAAyC,CAAC;AAC7D,mBAAmB,+CAA+C,CAAC;AACnE,YAAY,EACX,WAAW,EACX,4BAA4B,EAC5B,6BAA6B,EAC7B,oCAAoC,GACpC,MAAM,2CAA2C,CAAC;AAEnD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAE7D;;GAEG;AACH,eAAO,MAAM,qBAAqB,EAAE,MAAM,CAAC,YAAY,EAAE,eAAe,EAAE,CAgEzE,CAAC;AAoBF,eAAO,MAAM,UAAU,EAAE,MAAM,CAAC,YAAY,EAAE,QAAQ,GAAG,SAAS,CAwDxD,CAAC;AAEX,MAAM,WAAW,WAAW;IAC3B,WAAW,EAAE,MAAM,CAAC;IACpB,EAAE,EAAE,MAAM,CAAC;CACX;AAED,MAAM,MAAM,aAAa,GACtB;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,IAAI,EAAE,KAAK,CAAC;QACX,KAAK,EAAE,MAAM,CAAC;QACd,KAAK,EAAE,MAAM,CAAC;KACd,CAAC,CAAC;IACH,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,KAAK,CAAC;CACX,GACD;IACA,KAAK,EAAE,MAAM,EAAE,EAAE,CAAC;IAClB,IAAI,EAAE,SAAS,CAAC;CACf,GACD;IACA,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,MAAM,CAAC;CACZ,GACD;IACA,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;QACZ,KAAK,EAAE,MAAM,CAAC;QACd,IAAI,EAAE,MAAM,CAAC;KACb,CAAC,CAAC;IACH,IAAI,EAAE,kBAAkB,CAAC;CACxB,CAAC;AAEL,MAAM,WAAW,QAAQ;IACxB,MAAM,EAAE,aAAa,EAAE,CAAC;IACxB,OAAO,EAAE,aAAa,EAAE,CAAC;CACzB;AAED,MAAM,WAAW,QAAQ;IACxB,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,IAAI,EAAE,QAAQ,CAAC;IACf,EAAE,EAAE,YAAY,CAAC;IACjB,WAAW,CAAC,EAAE,YAAY,CAAC;IAC3B,aAAa,CAAC,EAAE,OAAO,CAAC;IACxB,KAAK,EAAE,MAAM,CAAC;IACd,SAAS,EAAE,eAAe,EAAE,CAAC;IAC7B,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,OAAO,EAAE,MAAM,CAAC;IAChB,YAAY,EAAE,MAAM,EAAE,CAAC;IACvB,SAAS,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,EAAE,IAAI,GAAG,OAAO,GAAG,WAAW,CAAC,CAAC"}
1
+ {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/tasks/index.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAC;AA8CpD,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,6CAA6C,CAAC;AACjE,YAAY,EACX,mBAAmB,EACnB,0BAA0B,EAC1B,oBAAoB,EACpB,4BAA4B,EAC5B,2BAA2B,EAC3B,0BAA0B,EAC1B,gCAAgC,EAChC,+BAA+B,GAC/B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,4CAA4C,CAAC;AAChE,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,0BAA0B,CAAC;AAC9C,YAAY,EACX,wBAAwB,EACxB,yBAAyB,EACzB,gCAAgC,EAChC,6BAA6B,GAC7B,MAAM,qCAAqC,CAAC;AAC7C,mBAAmB,+BAA+B,CAAC;AACnD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,oCAAoC,CAAC;AACxD,mBAAmB,8BAA8B,CAAC;AAClD,mBAAmB,yCAAyC,CAAC;AAC7D,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,sBAAsB,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,+BAA+B,CAAC;AACnH,mBAAmB,qCAAqC,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAC;AACtF,YAAY,EACX,6BAA6B,EAC7B,uBAAuB,EACvB,wBAAwB,EACxB,+BAA+B,EAC/B,4BAA4B,GAC5B,MAAM,oCAAoC,CAAC;AAC5C,YAAY,EACX,gCAAgC,EAChC,gCAAgC,EAChC,mBAAmB,EACnB,oBAAoB,EACpB,2BAA2B,EAC3B,qCAAqC,EACrC,kCAAkC,EAClC,yBAAyB,EACzB,uCAAuC,EACvC,0BAA0B,GAC1B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,0CAA0C,CAAC;AAC9D,mBAAmB,yCAAyC,CAAC;AAC7D,mBAAmB,+CAA+C,CAAC;AACnE,YAAY,EACX,WAAW,EACX,4BAA4B,EAC5B,6BAA6B,EAC7B,oCAAoC,GACpC,MAAM,2CAA2C,CAAC;AAEnD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAE7D;;GAEG;AACH,eAAO,MAAM,qBAAqB,EAAE,MAAM,CAAC,YAAY,EAAE,eAAe,EAAE,CAgEzE,CAAC;AAoBF,eAAO,MAAM,UAAU,EAAE,MAAM,CAAC,YAAY,EAAE,QAAQ,GAAG,SAAS,CAwDxD,CAAC;AAEX,MAAM,WAAW,WAAW;IAC3B,WAAW,EAAE,MAAM,CAAC;IACpB,EAAE,EAAE,MAAM,CAAC;CACX;AAED,MAAM,MAAM,aAAa,GACtB;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,IAAI,EAAE,KAAK,CAAC;QACX,KAAK,EAAE,MAAM,CAAC;QACd,KAAK,EAAE,MAAM,CAAC;KACd,CAAC,CAAC;IACH,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,KAAK,CAAC;CACX,GACD;IACA,KAAK,EAAE,MAAM,EAAE,EAAE,CAAC;IAClB,IAAI,EAAE,SAAS,CAAC;CACf,GACD;IACA,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,MAAM,CAAC;CACZ,GACD;IACA,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;QACZ,KAAK,EAAE,MAAM,CAAC;QACd,IAAI,EAAE,MAAM,CAAC;KACb,CAAC,CAAC;IACH,IAAI,EAAE,kBAAkB,CAAC;CACxB,CAAC;AAEL,MAAM,WAAW,QAAQ;IACxB,MAAM,EAAE,aAAa,EAAE,CAAC;IACxB,OAAO,EAAE,aAAa,EAAE,CAAC;CACzB;AAED,MAAM,WAAW,QAAQ;IACxB,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,IAAI,EAAE,QAAQ,CAAC;IACf,EAAE,EAAE,YAAY,CAAC;IACjB,WAAW,CAAC,EAAE,YAAY,CAAC;IAC3B,aAAa,CAAC,EAAE,OAAO,CAAC;IACxB,KAAK,EAAE,MAAM,CAAC;IACd,SAAS,EAAE,eAAe,EAAE,CAAC;IAC7B,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,OAAO,EAAE,MAAM,CAAC;IAChB,YAAY,EAAE,MAAM,EAAE,CAAC;IACvB,SAAS,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,EAAE,IAAI,GAAG,OAAO,GAAG,WAAW,CAAC,CAAC"}
@@ -5,47 +5,48 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
5
5
  Object.defineProperty(exports, "__esModule", { value: true });
6
6
  exports.TASKS_DATA = exports.TASKS_MODEL_LIBRARIES = void 0;
7
7
  const pipelines_js_1 = require("../pipelines.js");
8
- const data_js_1 = __importDefault(require("./audio-classification/data.js"));
9
- const data_js_2 = __importDefault(require("./audio-to-audio/data.js"));
10
- const data_js_3 = __importDefault(require("./automatic-speech-recognition/data.js"));
11
- const data_js_4 = __importDefault(require("./document-question-answering/data.js"));
12
- const data_js_5 = __importDefault(require("./feature-extraction/data.js"));
13
- const data_js_6 = __importDefault(require("./fill-mask/data.js"));
14
- const data_js_7 = __importDefault(require("./image-classification/data.js"));
15
- const data_js_8 = __importDefault(require("./image-feature-extraction/data.js"));
16
- const data_js_9 = __importDefault(require("./image-to-image/data.js"));
17
- const data_js_10 = __importDefault(require("./image-to-text/data.js"));
18
- const data_js_11 = __importDefault(require("./image-text-to-text/data.js"));
19
- const data_js_12 = __importDefault(require("./image-segmentation/data.js"));
20
- const data_js_13 = __importDefault(require("./mask-generation/data.js"));
21
- const data_js_14 = __importDefault(require("./object-detection/data.js"));
22
- const data_js_15 = __importDefault(require("./depth-estimation/data.js"));
23
- const data_js_16 = __importDefault(require("./placeholder/data.js"));
24
- const data_js_17 = __importDefault(require("./reinforcement-learning/data.js"));
25
- const data_js_18 = __importDefault(require("./question-answering/data.js"));
26
- const data_js_19 = __importDefault(require("./sentence-similarity/data.js"));
27
- const data_js_20 = __importDefault(require("./summarization/data.js"));
28
- const data_js_21 = __importDefault(require("./table-question-answering/data.js"));
29
- const data_js_22 = __importDefault(require("./tabular-classification/data.js"));
30
- const data_js_23 = __importDefault(require("./tabular-regression/data.js"));
31
- const data_js_24 = __importDefault(require("./text-to-image/data.js"));
32
- const data_js_25 = __importDefault(require("./text-to-speech/data.js"));
33
- const data_js_26 = __importDefault(require("./token-classification/data.js"));
34
- const data_js_27 = __importDefault(require("./translation/data.js"));
35
- const data_js_28 = __importDefault(require("./text-classification/data.js"));
36
- const data_js_29 = __importDefault(require("./text-generation/data.js"));
37
- const data_js_30 = __importDefault(require("./text-ranking/data.js"));
38
- const data_js_31 = __importDefault(require("./text-to-video/data.js"));
39
- const data_js_32 = __importDefault(require("./unconditional-image-generation/data.js"));
40
- const data_js_33 = __importDefault(require("./video-classification/data.js"));
41
- const data_js_34 = __importDefault(require("./visual-question-answering/data.js"));
42
- const data_js_35 = __importDefault(require("./zero-shot-classification/data.js"));
43
- const data_js_36 = __importDefault(require("./zero-shot-image-classification/data.js"));
44
- const data_js_37 = __importDefault(require("./zero-shot-object-detection/data.js"));
45
- const data_js_38 = __importDefault(require("./image-to-3d/data.js"));
46
- const data_js_39 = __importDefault(require("./text-to-3d/data.js"));
47
- const data_js_40 = __importDefault(require("./keypoint-detection/data.js"));
48
- const data_js_41 = __importDefault(require("./video-text-to-text/data.js"));
8
+ const data_js_1 = __importDefault(require("./any-to-any/data.js"));
9
+ const data_js_2 = __importDefault(require("./audio-classification/data.js"));
10
+ const data_js_3 = __importDefault(require("./audio-to-audio/data.js"));
11
+ const data_js_4 = __importDefault(require("./automatic-speech-recognition/data.js"));
12
+ const data_js_5 = __importDefault(require("./document-question-answering/data.js"));
13
+ const data_js_6 = __importDefault(require("./feature-extraction/data.js"));
14
+ const data_js_7 = __importDefault(require("./fill-mask/data.js"));
15
+ const data_js_8 = __importDefault(require("./image-classification/data.js"));
16
+ const data_js_9 = __importDefault(require("./image-feature-extraction/data.js"));
17
+ const data_js_10 = __importDefault(require("./image-to-image/data.js"));
18
+ const data_js_11 = __importDefault(require("./image-to-text/data.js"));
19
+ const data_js_12 = __importDefault(require("./image-text-to-text/data.js"));
20
+ const data_js_13 = __importDefault(require("./image-segmentation/data.js"));
21
+ const data_js_14 = __importDefault(require("./mask-generation/data.js"));
22
+ const data_js_15 = __importDefault(require("./object-detection/data.js"));
23
+ const data_js_16 = __importDefault(require("./depth-estimation/data.js"));
24
+ const data_js_17 = __importDefault(require("./placeholder/data.js"));
25
+ const data_js_18 = __importDefault(require("./reinforcement-learning/data.js"));
26
+ const data_js_19 = __importDefault(require("./question-answering/data.js"));
27
+ const data_js_20 = __importDefault(require("./sentence-similarity/data.js"));
28
+ const data_js_21 = __importDefault(require("./summarization/data.js"));
29
+ const data_js_22 = __importDefault(require("./table-question-answering/data.js"));
30
+ const data_js_23 = __importDefault(require("./tabular-classification/data.js"));
31
+ const data_js_24 = __importDefault(require("./tabular-regression/data.js"));
32
+ const data_js_25 = __importDefault(require("./text-to-image/data.js"));
33
+ const data_js_26 = __importDefault(require("./text-to-speech/data.js"));
34
+ const data_js_27 = __importDefault(require("./token-classification/data.js"));
35
+ const data_js_28 = __importDefault(require("./translation/data.js"));
36
+ const data_js_29 = __importDefault(require("./text-classification/data.js"));
37
+ const data_js_30 = __importDefault(require("./text-generation/data.js"));
38
+ const data_js_31 = __importDefault(require("./text-ranking/data.js"));
39
+ const data_js_32 = __importDefault(require("./text-to-video/data.js"));
40
+ const data_js_33 = __importDefault(require("./unconditional-image-generation/data.js"));
41
+ const data_js_34 = __importDefault(require("./video-classification/data.js"));
42
+ const data_js_35 = __importDefault(require("./visual-question-answering/data.js"));
43
+ const data_js_36 = __importDefault(require("./zero-shot-classification/data.js"));
44
+ const data_js_37 = __importDefault(require("./zero-shot-image-classification/data.js"));
45
+ const data_js_38 = __importDefault(require("./zero-shot-object-detection/data.js"));
46
+ const data_js_39 = __importDefault(require("./image-to-3d/data.js"));
47
+ const data_js_40 = __importDefault(require("./text-to-3d/data.js"));
48
+ const data_js_41 = __importDefault(require("./keypoint-detection/data.js"));
49
+ const data_js_42 = __importDefault(require("./video-text-to-text/data.js"));
49
50
  /**
50
51
  * Model libraries compatible with each ML task
51
52
  */
@@ -119,7 +120,7 @@ exports.TASKS_MODEL_LIBRARIES = {
119
120
  * If the partialTaskData argument is left undefined,
120
121
  * the default placholder data will be used.
121
122
  */
122
- function getData(type, partialTaskData = data_js_16.default) {
123
+ function getData(type, partialTaskData = data_js_17.default) {
123
124
  return {
124
125
  ...partialTaskData,
125
126
  id: type,
@@ -132,59 +133,59 @@ function getData(type, partialTaskData = data_js_16.default) {
132
133
  // Tasks that call getData() without the second argument will
133
134
  // have a "placeholder" page.
134
135
  exports.TASKS_DATA = {
135
- "any-to-any": getData("any-to-any", data_js_16.default),
136
- "audio-classification": getData("audio-classification", data_js_1.default),
137
- "audio-to-audio": getData("audio-to-audio", data_js_2.default),
138
- "audio-text-to-text": getData("audio-text-to-text", data_js_16.default),
139
- "automatic-speech-recognition": getData("automatic-speech-recognition", data_js_3.default),
140
- "depth-estimation": getData("depth-estimation", data_js_15.default),
141
- "document-question-answering": getData("document-question-answering", data_js_4.default),
142
- "visual-document-retrieval": getData("visual-document-retrieval", data_js_16.default),
143
- "feature-extraction": getData("feature-extraction", data_js_5.default),
144
- "fill-mask": getData("fill-mask", data_js_6.default),
136
+ "any-to-any": getData("any-to-any", data_js_1.default),
137
+ "audio-classification": getData("audio-classification", data_js_2.default),
138
+ "audio-to-audio": getData("audio-to-audio", data_js_3.default),
139
+ "audio-text-to-text": getData("audio-text-to-text", data_js_17.default),
140
+ "automatic-speech-recognition": getData("automatic-speech-recognition", data_js_4.default),
141
+ "depth-estimation": getData("depth-estimation", data_js_16.default),
142
+ "document-question-answering": getData("document-question-answering", data_js_5.default),
143
+ "visual-document-retrieval": getData("visual-document-retrieval", data_js_17.default),
144
+ "feature-extraction": getData("feature-extraction", data_js_6.default),
145
+ "fill-mask": getData("fill-mask", data_js_7.default),
145
146
  "graph-ml": undefined,
146
- "image-classification": getData("image-classification", data_js_7.default),
147
- "image-feature-extraction": getData("image-feature-extraction", data_js_8.default),
148
- "image-segmentation": getData("image-segmentation", data_js_12.default),
149
- "image-to-image": getData("image-to-image", data_js_9.default),
150
- "image-text-to-text": getData("image-text-to-text", data_js_11.default),
151
- "image-to-text": getData("image-to-text", data_js_10.default),
147
+ "image-classification": getData("image-classification", data_js_8.default),
148
+ "image-feature-extraction": getData("image-feature-extraction", data_js_9.default),
149
+ "image-segmentation": getData("image-segmentation", data_js_13.default),
150
+ "image-to-image": getData("image-to-image", data_js_10.default),
151
+ "image-text-to-text": getData("image-text-to-text", data_js_12.default),
152
+ "image-to-text": getData("image-to-text", data_js_11.default),
152
153
  "image-to-video": undefined,
153
- "keypoint-detection": getData("keypoint-detection", data_js_40.default),
154
- "mask-generation": getData("mask-generation", data_js_13.default),
154
+ "keypoint-detection": getData("keypoint-detection", data_js_41.default),
155
+ "mask-generation": getData("mask-generation", data_js_14.default),
155
156
  "multiple-choice": undefined,
156
- "object-detection": getData("object-detection", data_js_14.default),
157
- "video-classification": getData("video-classification", data_js_33.default),
157
+ "object-detection": getData("object-detection", data_js_15.default),
158
+ "video-classification": getData("video-classification", data_js_34.default),
158
159
  other: undefined,
159
- "question-answering": getData("question-answering", data_js_18.default),
160
- "reinforcement-learning": getData("reinforcement-learning", data_js_17.default),
160
+ "question-answering": getData("question-answering", data_js_19.default),
161
+ "reinforcement-learning": getData("reinforcement-learning", data_js_18.default),
161
162
  robotics: undefined,
162
- "sentence-similarity": getData("sentence-similarity", data_js_19.default),
163
- summarization: getData("summarization", data_js_20.default),
164
- "table-question-answering": getData("table-question-answering", data_js_21.default),
163
+ "sentence-similarity": getData("sentence-similarity", data_js_20.default),
164
+ summarization: getData("summarization", data_js_21.default),
165
+ "table-question-answering": getData("table-question-answering", data_js_22.default),
165
166
  "table-to-text": undefined,
166
- "tabular-classification": getData("tabular-classification", data_js_22.default),
167
- "tabular-regression": getData("tabular-regression", data_js_23.default),
167
+ "tabular-classification": getData("tabular-classification", data_js_23.default),
168
+ "tabular-regression": getData("tabular-regression", data_js_24.default),
168
169
  "tabular-to-text": undefined,
169
- "text-classification": getData("text-classification", data_js_28.default),
170
- "text-generation": getData("text-generation", data_js_29.default),
171
- "text-ranking": getData("text-ranking", data_js_30.default),
170
+ "text-classification": getData("text-classification", data_js_29.default),
171
+ "text-generation": getData("text-generation", data_js_30.default),
172
+ "text-ranking": getData("text-ranking", data_js_31.default),
172
173
  "text-retrieval": undefined,
173
- "text-to-image": getData("text-to-image", data_js_24.default),
174
- "text-to-speech": getData("text-to-speech", data_js_25.default),
174
+ "text-to-image": getData("text-to-image", data_js_25.default),
175
+ "text-to-speech": getData("text-to-speech", data_js_26.default),
175
176
  "text-to-audio": undefined,
176
- "text-to-video": getData("text-to-video", data_js_31.default),
177
+ "text-to-video": getData("text-to-video", data_js_32.default),
177
178
  "text2text-generation": undefined,
178
179
  "time-series-forecasting": undefined,
179
- "token-classification": getData("token-classification", data_js_26.default),
180
- translation: getData("translation", data_js_27.default),
181
- "unconditional-image-generation": getData("unconditional-image-generation", data_js_32.default),
182
- "video-text-to-text": getData("video-text-to-text", data_js_41.default),
183
- "visual-question-answering": getData("visual-question-answering", data_js_34.default),
180
+ "token-classification": getData("token-classification", data_js_27.default),
181
+ translation: getData("translation", data_js_28.default),
182
+ "unconditional-image-generation": getData("unconditional-image-generation", data_js_33.default),
183
+ "video-text-to-text": getData("video-text-to-text", data_js_42.default),
184
+ "visual-question-answering": getData("visual-question-answering", data_js_35.default),
184
185
  "voice-activity-detection": undefined,
185
- "zero-shot-classification": getData("zero-shot-classification", data_js_35.default),
186
- "zero-shot-image-classification": getData("zero-shot-image-classification", data_js_36.default),
187
- "zero-shot-object-detection": getData("zero-shot-object-detection", data_js_37.default),
188
- "text-to-3d": getData("text-to-3d", data_js_39.default),
189
- "image-to-3d": getData("image-to-3d", data_js_38.default),
186
+ "zero-shot-classification": getData("zero-shot-classification", data_js_36.default),
187
+ "zero-shot-image-classification": getData("zero-shot-image-classification", data_js_37.default),
188
+ "zero-shot-object-detection": getData("zero-shot-object-detection", data_js_38.default),
189
+ "text-to-3d": getData("text-to-3d", data_js_40.default),
190
+ "image-to-3d": getData("image-to-3d", data_js_39.default),
190
191
  };
@@ -430,6 +430,18 @@ export declare const SKUS: {
430
430
  tflops: number;
431
431
  memory: number[];
432
432
  };
433
+ "RX 6650 XT": {
434
+ tflops: number;
435
+ memory: number[];
436
+ };
437
+ "RX 6600 XT": {
438
+ tflops: number;
439
+ memory: number[];
440
+ };
441
+ "RX 6600": {
442
+ tflops: number;
443
+ memory: number[];
444
+ };
433
445
  "Radeon Pro VII": {
434
446
  tflops: number;
435
447
  memory: number[];
@@ -1 +1 @@
1
- {"version":3,"file":"hardware.d.ts","sourceRoot":"","sources":["../../src/hardware.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,eAAO,MAAM,iDAAiD,QAAW,CAAC;AAC1E,eAAO,MAAM,yDAAyD,QAAW,CAAC;AAClF,eAAO,MAAM,oCAAoC,QAAU,CAAC;AAE5D;;;GAGG;AACH,eAAO,MAAM,+CAA+C,QAAW,CAAC;AAExE,MAAM,WAAW,YAAY;IAC5B;;;;;;;;;OASG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,EAAE,CAAC;CAClB;AAED,eAAO,MAAM,sBAAsB,UAAqD,CAAC;AAEzF,eAAO,MAAM,IAAI;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA4mBuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}
1
+ {"version":3,"file":"hardware.d.ts","sourceRoot":"","sources":["../../src/hardware.ts"],"names":[],"mappings":"AAAA;;;GAGG;AACH,eAAO,MAAM,iDAAiD,QAAW,CAAC;AAC1E,eAAO,MAAM,yDAAyD,QAAW,CAAC;AAClF,eAAO,MAAM,oCAAoC,QAAU,CAAC;AAE5D;;;GAGG;AACH,eAAO,MAAM,+CAA+C,QAAW,CAAC;AAExE,MAAM,WAAW,YAAY;IAC5B;;;;;;;;;OASG;IACH,MAAM,EAAE,MAAM,CAAC;IACf;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,EAAE,CAAC;CAClB;AAED,eAAO,MAAM,sBAAsB,UAAqD,CAAC;AAEzF,eAAO,MAAM,IAAI;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAwnBuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}
@@ -412,6 +412,18 @@ export const SKUS = {
412
412
  tflops: 22.58,
413
413
  memory: [10],
414
414
  },
415
+ "RX 6650 XT": {
416
+ tflops: 21.59,
417
+ memory: [8],
418
+ },
419
+ "RX 6600 XT": {
420
+ tflops: 21.21,
421
+ memory: [8],
422
+ },
423
+ "RX 6600": {
424
+ tflops: 17.86,
425
+ memory: [8],
426
+ },
415
427
  "Radeon Pro VII": {
416
428
  tflops: 26.11,
417
429
  memory: [16],
@@ -0,0 +1,4 @@
1
+ import type { TaskDataCustom } from "../index.js";
2
+ declare const taskData: TaskDataCustom;
3
+ export default taskData;
4
+ //# sourceMappingURL=data.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"data.d.ts","sourceRoot":"","sources":["../../../../src/tasks/any-to-any/data.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,aAAa,CAAC;AAElD,QAAA,MAAM,QAAQ,EAAE,cA6Df,CAAC;AAEF,eAAe,QAAQ,CAAC"}
@@ -0,0 +1,61 @@
1
+ const taskData = {
2
+ datasets: [
3
+ {
4
+ description: "A dataset with multiple modality input and output pairs.",
5
+ id: "PKU-Alignment/align-anything",
6
+ },
7
+ ],
8
+ demo: {
9
+ inputs: [
10
+ {
11
+ filename: "any-to-any-input.jpg",
12
+ type: "img",
13
+ },
14
+ {
15
+ label: "Text Prompt",
16
+ content: "Describe the position of the bee in detail.",
17
+ type: "text",
18
+ },
19
+ ],
20
+ outputs: [
21
+ {
22
+ label: "Generated Text",
23
+ content: "The place in the picture is Osaka Castle, located in Osaka, Japan. Osaka Castle is a historic castle that was originally built in the 16th century by Toyotomi Hideyoshi, a powerful warlord of the time. It is one of the most famous landmarks in Osaka and is known for its distinctive white walls and black roof tiles. The castle has been rebuilt several times over the centuries and is now a popular tourist attraction, offering visitors a glimpse into Japan's rich history and culture.",
24
+ type: "text",
25
+ },
26
+ {
27
+ filename: "any-to-any-output.wav",
28
+ type: "audio",
29
+ },
30
+ ],
31
+ },
32
+ metrics: [],
33
+ models: [
34
+ {
35
+ description: "Strong model that can take in video, audio, image, text and output text and natural speech.",
36
+ id: "Qwen/Qwen2.5-Omni-7B",
37
+ },
38
+ {
39
+ description: "Robust model that can take in image and text and generate image and text.",
40
+ id: "deepseek-ai/Janus-Pro-7B",
41
+ },
42
+ {
43
+ description: "Any-to-any model with speech, video, audio, image and text understanding capabilities.",
44
+ id: "openbmb/MiniCPM-o-2_6",
45
+ },
46
+ {
47
+ description: "A model that can understand image and text and generate image and text.",
48
+ id: "EPFL-VILAB/4M-21_XL",
49
+ },
50
+ ],
51
+ spaces: [
52
+ {
53
+ description: "An application to chat with an any-to-any (image & text) model.",
54
+ id: "deepseek-ai/Janus-Pro-7B",
55
+ },
56
+ ],
57
+ summary: "Any-to-any models can understand two or more modalities and output two or more modalities.",
58
+ widgetModels: [],
59
+ youtubeId: "",
60
+ };
61
+ export default taskData;
@@ -1 +1 @@
1
- {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/tasks/index.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAC;AA6CpD,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,6CAA6C,CAAC;AACjE,YAAY,EACX,mBAAmB,EACnB,0BAA0B,EAC1B,oBAAoB,EACpB,4BAA4B,EAC5B,2BAA2B,EAC3B,0BAA0B,EAC1B,gCAAgC,EAChC,+BAA+B,GAC/B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,4CAA4C,CAAC;AAChE,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,0BAA0B,CAAC;AAC9C,YAAY,EACX,wBAAwB,EACxB,yBAAyB,EACzB,gCAAgC,EAChC,6BAA6B,GAC7B,MAAM,qCAAqC,CAAC;AAC7C,mBAAmB,+BAA+B,CAAC;AACnD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,oCAAoC,CAAC;AACxD,mBAAmB,8BAA8B,CAAC;AAClD,mBAAmB,yCAAyC,CAAC;AAC7D,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,sBAAsB,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,+BAA+B,CAAC;AACnH,mBAAmB,qCAAqC,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAC;AACtF,YAAY,EACX,6BAA6B,EAC7B,uBAAuB,EACvB,wBAAwB,EACxB,+BAA+B,EAC/B,4BAA4B,GAC5B,MAAM,oCAAoC,CAAC;AAC5C,YAAY,EACX,gCAAgC,EAChC,gCAAgC,EAChC,mBAAmB,EACnB,oBAAoB,EACpB,2BAA2B,EAC3B,qCAAqC,EACrC,kCAAkC,EAClC,yBAAyB,EACzB,uCAAuC,EACvC,0BAA0B,GAC1B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,0CAA0C,CAAC;AAC9D,mBAAmB,yCAAyC,CAAC;AAC7D,mBAAmB,+CAA+C,CAAC;AACnE,YAAY,EACX,WAAW,EACX,4BAA4B,EAC5B,6BAA6B,EAC7B,oCAAoC,GACpC,MAAM,2CAA2C,CAAC;AAEnD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAE7D;;GAEG;AACH,eAAO,MAAM,qBAAqB,EAAE,MAAM,CAAC,YAAY,EAAE,eAAe,EAAE,CAgEzE,CAAC;AAoBF,eAAO,MAAM,UAAU,EAAE,MAAM,CAAC,YAAY,EAAE,QAAQ,GAAG,SAAS,CAwDxD,CAAC;AAEX,MAAM,WAAW,WAAW;IAC3B,WAAW,EAAE,MAAM,CAAC;IACpB,EAAE,EAAE,MAAM,CAAC;CACX;AAED,MAAM,MAAM,aAAa,GACtB;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,IAAI,EAAE,KAAK,CAAC;QACX,KAAK,EAAE,MAAM,CAAC;QACd,KAAK,EAAE,MAAM,CAAC;KACd,CAAC,CAAC;IACH,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,KAAK,CAAC;CACX,GACD;IACA,KAAK,EAAE,MAAM,EAAE,EAAE,CAAC;IAClB,IAAI,EAAE,SAAS,CAAC;CACf,GACD;IACA,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,MAAM,CAAC;CACZ,GACD;IACA,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;QACZ,KAAK,EAAE,MAAM,CAAC;QACd,IAAI,EAAE,MAAM,CAAC;KACb,CAAC,CAAC;IACH,IAAI,EAAE,kBAAkB,CAAC;CACxB,CAAC;AAEL,MAAM,WAAW,QAAQ;IACxB,MAAM,EAAE,aAAa,EAAE,CAAC;IACxB,OAAO,EAAE,aAAa,EAAE,CAAC;CACzB;AAED,MAAM,WAAW,QAAQ;IACxB,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,IAAI,EAAE,QAAQ,CAAC;IACf,EAAE,EAAE,YAAY,CAAC;IACjB,WAAW,CAAC,EAAE,YAAY,CAAC;IAC3B,aAAa,CAAC,EAAE,OAAO,CAAC;IACxB,KAAK,EAAE,MAAM,CAAC;IACd,SAAS,EAAE,eAAe,EAAE,CAAC;IAC7B,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,OAAO,EAAE,MAAM,CAAC;IAChB,YAAY,EAAE,MAAM,EAAE,CAAC;IACvB,SAAS,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,EAAE,IAAI,GAAG,OAAO,GAAG,WAAW,CAAC,CAAC"}
1
+ {"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../../../src/tasks/index.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,iBAAiB,CAAC;AA8CpD,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,6CAA6C,CAAC;AACjE,YAAY,EACX,mBAAmB,EACnB,0BAA0B,EAC1B,oBAAoB,EACpB,4BAA4B,EAC5B,2BAA2B,EAC3B,0BAA0B,EAC1B,gCAAgC,EAChC,+BAA+B,GAC/B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,4CAA4C,CAAC;AAChE,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,0BAA0B,CAAC;AAC9C,YAAY,EACX,wBAAwB,EACxB,yBAAyB,EACzB,gCAAgC,EAChC,6BAA6B,GAC7B,MAAM,qCAAqC,CAAC;AAC7C,mBAAmB,+BAA+B,CAAC;AACnD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,iCAAiC,CAAC;AACrD,mBAAmB,mCAAmC,CAAC;AACvD,mBAAmB,oCAAoC,CAAC;AACxD,mBAAmB,8BAA8B,CAAC;AAClD,mBAAmB,yCAAyC,CAAC;AAC7D,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,qBAAqB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,qBAAqB,EAAE,iBAAiB,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAC/G,YAAY,EAAE,sBAAsB,EAAE,iBAAiB,EAAE,kBAAkB,EAAE,MAAM,+BAA+B,CAAC;AACnH,mBAAmB,qCAAqC,CAAC;AACzD,YAAY,EAAE,gBAAgB,EAAE,iBAAiB,EAAE,MAAM,4BAA4B,CAAC;AACtF,YAAY,EACX,6BAA6B,EAC7B,uBAAuB,EACvB,wBAAwB,EACxB,+BAA+B,EAC/B,4BAA4B,GAC5B,MAAM,oCAAoC,CAAC;AAC5C,YAAY,EACX,gCAAgC,EAChC,gCAAgC,EAChC,mBAAmB,EACnB,oBAAoB,EACpB,2BAA2B,EAC3B,qCAAqC,EACrC,kCAAkC,EAClC,yBAAyB,EACzB,uCAAuC,EACvC,0BAA0B,GAC1B,MAAM,gCAAgC,CAAC;AACxC,mBAAmB,qCAAqC,CAAC;AACzD,mBAAmB,0CAA0C,CAAC;AAC9D,mBAAmB,yCAAyC,CAAC;AAC7D,mBAAmB,+CAA+C,CAAC;AACnE,YAAY,EACX,WAAW,EACX,4BAA4B,EAC5B,6BAA6B,EAC7B,oCAAoC,GACpC,MAAM,2CAA2C,CAAC;AAEnD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAE7D;;GAEG;AACH,eAAO,MAAM,qBAAqB,EAAE,MAAM,CAAC,YAAY,EAAE,eAAe,EAAE,CAgEzE,CAAC;AAoBF,eAAO,MAAM,UAAU,EAAE,MAAM,CAAC,YAAY,EAAE,QAAQ,GAAG,SAAS,CAwDxD,CAAC;AAEX,MAAM,WAAW,WAAW;IAC3B,WAAW,EAAE,MAAM,CAAC;IACpB,EAAE,EAAE,MAAM,CAAC;CACX;AAED,MAAM,MAAM,aAAa,GACtB;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,IAAI,EAAE,KAAK,CAAC;QACX,KAAK,EAAE,MAAM,CAAC;QACd,KAAK,EAAE,MAAM,CAAC;KACd,CAAC,CAAC;IACH,IAAI,EAAE,OAAO,CAAC;CACb,GACD;IACA,QAAQ,EAAE,MAAM,CAAC;IACjB,IAAI,EAAE,KAAK,CAAC;CACX,GACD;IACA,KAAK,EAAE,MAAM,EAAE,EAAE,CAAC;IAClB,IAAI,EAAE,SAAS,CAAC;CACf,GACD;IACA,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,MAAM,CAAC;IACd,IAAI,EAAE,MAAM,CAAC;CACZ,GACD;IACA,IAAI,EAAE,MAAM,CAAC;IACb,MAAM,EAAE,KAAK,CAAC;QACb,GAAG,EAAE,MAAM,CAAC;QACZ,KAAK,EAAE,MAAM,CAAC;QACd,IAAI,EAAE,MAAM,CAAC;KACb,CAAC,CAAC;IACH,IAAI,EAAE,kBAAkB,CAAC;CACxB,CAAC;AAEL,MAAM,WAAW,QAAQ;IACxB,MAAM,EAAE,aAAa,EAAE,CAAC;IACxB,OAAO,EAAE,aAAa,EAAE,CAAC;CACzB;AAED,MAAM,WAAW,QAAQ;IACxB,QAAQ,EAAE,WAAW,EAAE,CAAC;IACxB,IAAI,EAAE,QAAQ,CAAC;IACf,EAAE,EAAE,YAAY,CAAC;IACjB,WAAW,CAAC,EAAE,YAAY,CAAC;IAC3B,aAAa,CAAC,EAAE,OAAO,CAAC;IACxB,KAAK,EAAE,MAAM,CAAC;IACd,SAAS,EAAE,eAAe,EAAE,CAAC;IAC7B,OAAO,EAAE,WAAW,EAAE,CAAC;IACvB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,MAAM,EAAE,WAAW,EAAE,CAAC;IACtB,OAAO,EAAE,MAAM,CAAC;IAChB,YAAY,EAAE,MAAM,EAAE,CAAC;IACvB,SAAS,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,MAAM,cAAc,GAAG,IAAI,CAAC,QAAQ,EAAE,IAAI,GAAG,OAAO,GAAG,WAAW,CAAC,CAAC"}
@@ -1,4 +1,5 @@
1
1
  import { PIPELINE_DATA } from "../pipelines.js";
2
+ import anyToAny from "./any-to-any/data.js";
2
3
  import audioClassification from "./audio-classification/data.js";
3
4
  import audioToAudio from "./audio-to-audio/data.js";
4
5
  import automaticSpeechRecognition from "./automatic-speech-recognition/data.js";
@@ -126,7 +127,7 @@ function getData(type, partialTaskData = placeholder) {
126
127
  // Tasks that call getData() without the second argument will
127
128
  // have a "placeholder" page.
128
129
  export const TASKS_DATA = {
129
- "any-to-any": getData("any-to-any", placeholder),
130
+ "any-to-any": getData("any-to-any", anyToAny),
130
131
  "audio-classification": getData("audio-classification", audioClassification),
131
132
  "audio-to-audio": getData("audio-to-audio", audioToAudio),
132
133
  "audio-text-to-text": getData("audio-text-to-text", placeholder),
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@huggingface/tasks",
3
3
  "packageManager": "pnpm@8.10.5",
4
- "version": "0.18.7",
4
+ "version": "0.18.8",
5
5
  "description": "List of ML tasks for huggingface.co/tasks",
6
6
  "repository": "https://github.com/huggingface/huggingface.js.git",
7
7
  "publishConfig": {
package/src/hardware.ts CHANGED
@@ -434,6 +434,18 @@ export const SKUS = {
434
434
  tflops: 22.58,
435
435
  memory: [10],
436
436
  },
437
+ "RX 6650 XT": {
438
+ tflops: 21.59,
439
+ memory: [8],
440
+ },
441
+ "RX 6600 XT": {
442
+ tflops: 21.21,
443
+ memory: [8],
444
+ },
445
+ "RX 6600": {
446
+ tflops: 17.86,
447
+ memory: [8],
448
+ },
437
449
  "Radeon Pro VII": {
438
450
  tflops: 26.11,
439
451
  memory: [16],
@@ -0,0 +1,58 @@
1
+ ## Use Cases
2
+
3
+ ### Embodied Agents
4
+
5
+ Any-to-any models can help embodied agents operate in multi-sensory environments, such as video games or physical robots. The model can take in an image or video of a scene, text prompts, and audio, and respond by generating text, actions, predict next frames, or generate speech commands.
6
+
7
+ ### Real-time Accessibility Systems
8
+
9
+ Vision-language based any-to-any models can be used aid visually impaired people. A real-time on-device any-to-any model can take a real-world video stream from wearable glasses, and describe the scene in audio (e.g., "A person in a red coat is walking toward you") or provide real-time closed captions and environmental sound cues.
10
+
11
+ ### Multimodal Content Creation
12
+
13
+ One can use any-to-any models to generate multimodal content. For example, given a video and an outline, the model can generate speech, better videos, or a descriptive blog post. Moreover, these models can sync narration timing with visual transitions.
14
+
15
+ ## Inference
16
+
17
+ You can infer with any-to-any models using transformers. Below is an example to infer Qwen2.5-Omni-7B model, make sure to check the model you're inferring with.
18
+
19
+ ```python
20
+ import soundfile as sf
21
+ from transformers import Qwen2_5OmniModel, Qwen2_5OmniProcessor
22
+ from qwen_omni_utils import process_mm_info
23
+
24
+ model = Qwen2_5OmniModel.from_pretrained("Qwen/Qwen2.5-Omni-7B", torch_dtype="auto", device_map="auto")
25
+
26
+ processor = Qwen2_5OmniProcessor.from_pretrained("Qwen/Qwen2.5-Omni-7B")
27
+
28
+ conversation = [
29
+ {
30
+ "role": "system",
31
+ "content": "You are Qwen, a virtual human developed by the Qwen Team, Alibaba Group, capable of perceiving auditory and visual inputs, as well as generating text and speech.",
32
+ },
33
+ {
34
+ "role": "user",
35
+ "content": [
36
+ {"type": "video", "video": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen2.5-Omni/draw.mp4"},
37
+ ],
38
+ },
39
+ ]
40
+
41
+ USE_AUDIO_IN_VIDEO = True
42
+
43
+ text = processor.apply_chat_template(conversation, add_generation_prompt=True, tokenize=False)
44
+ audios, images, videos = process_mm_info(conversation, use_audio_in_video=USE_AUDIO_IN_VIDEO)
45
+ inputs = processor(text=text, audios=audios, images=images, videos=videos, return_tensors="pt", padding=True, use_audio_in_video=USE_AUDIO_IN_VIDEO)
46
+ inputs = inputs.to(model.device).to(model.dtype)
47
+
48
+ # Inference: Generation of the output text and audio
49
+ text_ids, audio = model.generate(**inputs, use_audio_in_video=USE_AUDIO_IN_VIDEO)
50
+
51
+ text = processor.batch_decode(text_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)
52
+ print(text)
53
+ sf.write(
54
+ "output.wav",
55
+ audio.reshape(-1).detach().cpu().numpy(),
56
+ samplerate=24000,
57
+ )
58
+ ```
@@ -0,0 +1,66 @@
1
+ import type { TaskDataCustom } from "../index.js";
2
+
3
+ const taskData: TaskDataCustom = {
4
+ datasets: [
5
+ {
6
+ description: "A dataset with multiple modality input and output pairs.",
7
+ id: "PKU-Alignment/align-anything",
8
+ },
9
+ ],
10
+ demo: {
11
+ inputs: [
12
+ {
13
+ filename: "any-to-any-input.jpg",
14
+ type: "img",
15
+ },
16
+ {
17
+ label: "Text Prompt",
18
+ content: "Describe the position of the bee in detail.",
19
+ type: "text",
20
+ },
21
+ ],
22
+ outputs: [
23
+ {
24
+ label: "Generated Text",
25
+ content:
26
+ "The place in the picture is Osaka Castle, located in Osaka, Japan. Osaka Castle is a historic castle that was originally built in the 16th century by Toyotomi Hideyoshi, a powerful warlord of the time. It is one of the most famous landmarks in Osaka and is known for its distinctive white walls and black roof tiles. The castle has been rebuilt several times over the centuries and is now a popular tourist attraction, offering visitors a glimpse into Japan's rich history and culture.",
27
+ type: "text",
28
+ },
29
+ {
30
+ filename: "any-to-any-output.wav",
31
+ type: "audio",
32
+ },
33
+ ],
34
+ },
35
+ metrics: [],
36
+ models: [
37
+ {
38
+ description: "Strong model that can take in video, audio, image, text and output text and natural speech.",
39
+ id: "Qwen/Qwen2.5-Omni-7B",
40
+ },
41
+ {
42
+ description: "Robust model that can take in image and text and generate image and text.",
43
+ id: "deepseek-ai/Janus-Pro-7B",
44
+ },
45
+ {
46
+ description: "Any-to-any model with speech, video, audio, image and text understanding capabilities.",
47
+ id: "openbmb/MiniCPM-o-2_6",
48
+ },
49
+ {
50
+ description: "A model that can understand image and text and generate image and text.",
51
+ id: "EPFL-VILAB/4M-21_XL",
52
+ },
53
+ ],
54
+ spaces: [
55
+ {
56
+ description: "An application to chat with an any-to-any (image & text) model.",
57
+ id: "deepseek-ai/Janus-Pro-7B",
58
+ },
59
+ ],
60
+ summary:
61
+ "Any-to-any models can understand two or more modalities and output two or more modalities.",
62
+ widgetModels: [],
63
+ youtubeId: "",
64
+ };
65
+
66
+ export default taskData;
@@ -1,6 +1,7 @@
1
1
  import type { PipelineType } from "../pipelines.js";
2
2
  import { PIPELINE_DATA } from "../pipelines.js";
3
3
 
4
+ import anyToAny from "./any-to-any/data.js";
4
5
  import audioClassification from "./audio-classification/data.js";
5
6
  import audioToAudio from "./audio-to-audio/data.js";
6
7
  import automaticSpeechRecognition from "./automatic-speech-recognition/data.js";
@@ -198,7 +199,7 @@ function getData(type: PipelineType, partialTaskData: TaskDataCustom = placehold
198
199
  // Tasks that call getData() without the second argument will
199
200
  // have a "placeholder" page.
200
201
  export const TASKS_DATA: Record<PipelineType, TaskData | undefined> = {
201
- "any-to-any": getData("any-to-any", placeholder),
202
+ "any-to-any": getData("any-to-any", anyToAny),
202
203
  "audio-classification": getData("audio-classification", audioClassification),
203
204
  "audio-to-audio": getData("audio-to-audio", audioToAudio),
204
205
  "audio-text-to-text": getData("audio-text-to-text", placeholder),