@huggingface/tasks 0.10.8 → 0.10.9

package/dist/index.cjs

@@ -1541,16 +1541,16 @@ var data_default2 = taskData2;
 var taskData3 = {
   datasets: [
     {
-      description: "18,000 hours of multilingual audio-text dataset in 108 languages.",
-      id: "mozilla-foundation/common_voice_13_0"
+      description: "31,175 hours of multilingual audio-text dataset in 108 languages.",
+      id: "mozilla-foundation/common_voice_17_0"
     },
     {
       description: "An English dataset with 1,000 hours of data.",
       id: "librispeech_asr"
     },
     {
-      description: "High quality, multi-speaker audio data and their transcriptions in various languages.",
-      id: "openslr"
+      description: "A multi-lingual audio dataset with 370K hours of audio.",
+      id: "espnet/yodas"
     }
   ],
   demo: {
@@ -1585,12 +1585,12 @@ var taskData3 = {
       id: "openai/whisper-large-v3"
     },
     {
-      description: "A good generic ASR model by MetaAI.",
-      id: "facebook/wav2vec2-base-960h"
+      description: "A good generic speech model by MetaAI for fine-tuning.",
+      id: "facebook/w2v-bert-2.0"
     },
     {
       description: "An end-to-end model that performs ASR and Speech Translation by MetaAI.",
-      id: "facebook/s2t-small-mustc-en-fr-st"
+      id: "facebook/seamless-m4t-v2-large"
     }
   ],
   spaces: [
@@ -2992,8 +2992,8 @@ var taskData24 = {
   canonicalId: "text-to-audio",
   datasets: [
     {
-      description: "Thousands of short audio clips of a single speaker.",
-      id: "lj_speech"
+      description: "10K hours of multi-speaker English dataset.",
+      id: "parler-tts/mls_eng_10k"
     },
     {
       description: "Multi-speaker English dataset.",
@@ -3031,8 +3031,8 @@ var taskData24 = {
       id: "facebook/mms-tts"
     },
     {
-      description: "An end-to-end speech synthesis model.",
-      id: "microsoft/speecht5_tts"
+      description: "A prompt based, powerful TTS model.",
+      id: "parler-tts/parler_tts_mini_v0.1"
     }
   ],
   spaces: [
@@ -3045,8 +3045,8 @@ var taskData24 = {
       id: "coqui/xtts"
     },
     {
-      description: "An application that synthesizes speech for various speaker types.",
-      id: "Matthijs/speecht5-tts-demo"
+      description: "An application that synthesizes speech for diverse speaker prompts.",
+      id: "parler-tts/parler_tts_mini"
     }
   ],
   summary: "Text-to-Speech (TTS) is the task of generating natural sounding speech given text input. TTS models can be extended to have a single model that generates speech for multiple speakers and multiple languages.",
@@ -5476,6 +5476,30 @@ var SKUS = {
         tflops: 91.1,
         memory: [48]
       },
+      "RTX 5880 Ada": {
+        tflops: 69.3,
+        memory: [48]
+      },
+      "RTX 5000 Ada": {
+        tflops: 65.3,
+        memory: [32]
+      },
+      "RTX 4500 Ada": {
+        tflops: 39.6,
+        memory: [24]
+      },
+      "RTX 4000 Ada": {
+        tflops: 26.7,
+        memory: [20]
+      },
+      "RTX 4000 SFF Ada": {
+        tflops: 19.2,
+        memory: [20]
+      },
+      "RTX 2000 Ada": {
+        tflops: 12,
+        memory: [16]
+      },
       A100: {
         tflops: 77.97,
         memory: [80, 40]
@@ -5488,14 +5512,14 @@ var SKUS = {
         tflops: 31.24,
         memory: [24]
       },
-      T4: {
-        tflops: 65.13,
-        memory: [16]
-      },
       "RTX 4090": {
         tflops: 82.58,
         memory: [24]
       },
+      "RTX 4090D": {
+        tflops: 79.49,
+        memory: [24]
+      },
       "RTX 4080 SUPER": {
         tflops: 52.2,
         memory: [16]
@@ -5520,6 +5544,14 @@ var SKUS = {
         tflops: 44.1,
         memory: [16]
       },
+      "RTX 4060": {
+        tflops: 15.11,
+        memory: [8]
+      },
+      "RTX 4060 Ti": {
+        tflops: 22.06,
+        memory: [8, 16]
+      },
       "RTX 3090": {
         tflops: 35.58,
         memory: [24]
@@ -5556,13 +5588,53 @@ var SKUS = {
         tflops: 12.74,
         memory: [12, 8]
       },
+      "RTX 2070": {
+        tflops: 14.93,
+        memory: [8]
+      },
+      "RTX 3050 Mobile": {
+        tflops: 7.639,
+        memory: [6]
+      },
       "RTX 2060 Mobile": {
         tflops: 9.22,
         memory: [6]
       },
+      "GTX 1080 Ti": {
+        tflops: 11.34,
+        // float32 (GPU does not support native float16)
+        memory: [11]
+      },
+      "GTX 1070 Ti": {
+        tflops: 8.2,
+        // float32 (GPU does not support native float16)
+        memory: [8]
+      },
       "RTX Titan": {
         tflops: 32.62,
         memory: [24]
+      },
+      "GTX 1650 Mobile": {
+        tflops: 6.39,
+        memory: [4]
+      },
+      T4: {
+        tflops: 65.13,
+        memory: [16]
+      },
+      V100: {
+        tflops: 28.26,
+        memory: [32, 16]
+      },
+      "Quadro P6000": {
+        tflops: 12.63,
+        // float32 (GPU does not support native float16)
+        memory: [24]
+      },
+      P40: {
+        tflops: 11.76,
+        // float32 (GPU does not support native float16)
+        memory: [24]
       }
     },
     AMD: {
@@ -5578,6 +5650,10 @@ var SKUS = {
         tflops: 181,
         memory: [64]
       },
+      MI100: {
+        tflops: 184.6,
+        memory: [32]
+      },
       "RX 7900 XTX": {
         tflops: 122.8,
         memory: [24]
@@ -5601,6 +5677,18 @@ var SKUS = {
       "RX 7600 XT": {
         tflops: 45.14,
         memory: [16, 8]
+      },
+      "RX 6950 XT": {
+        tflops: 47.31,
+        memory: [16]
+      },
+      "RX 6800": {
+        tflops: 32.33,
+        memory: [16]
+      },
+      "Radeon Pro VII": {
+        tflops: 26.11,
+        memory: [16]
       }
     }
   },
@@ -5811,6 +5899,13 @@ var LOCAL_APPS = {
     displayOnModelPage: isGgufModel,
     deeplink: (model) => new URL(`https://backyard.ai/hf/model/${model.id}`)
   },
+  sanctum: {
+    prettyLabel: "Sanctum",
+    docsUrl: "https://sanctum.ai",
+    mainTask: "text-generation",
+    displayOnModelPage: isGgufModel,
+    deeplink: (model) => new URL(`sanctum://open_from_hf?model=${model.id}`)
+  },
   drawthings: {
     prettyLabel: "Draw Things",
     docsUrl: "https://drawthings.ai",

package/dist/index.js

@@ -1503,16 +1503,16 @@ var data_default2 = taskData2;
 var taskData3 = {
   datasets: [
     {
-      description: "18,000 hours of multilingual audio-text dataset in 108 languages.",
-      id: "mozilla-foundation/common_voice_13_0"
+      description: "31,175 hours of multilingual audio-text dataset in 108 languages.",
+      id: "mozilla-foundation/common_voice_17_0"
     },
     {
       description: "An English dataset with 1,000 hours of data.",
       id: "librispeech_asr"
     },
     {
-      description: "High quality, multi-speaker audio data and their transcriptions in various languages.",
-      id: "openslr"
+      description: "A multi-lingual audio dataset with 370K hours of audio.",
+      id: "espnet/yodas"
     }
   ],
   demo: {
@@ -1547,12 +1547,12 @@ var taskData3 = {
       id: "openai/whisper-large-v3"
     },
     {
-      description: "A good generic ASR model by MetaAI.",
-      id: "facebook/wav2vec2-base-960h"
+      description: "A good generic speech model by MetaAI for fine-tuning.",
+      id: "facebook/w2v-bert-2.0"
     },
     {
       description: "An end-to-end model that performs ASR and Speech Translation by MetaAI.",
-      id: "facebook/s2t-small-mustc-en-fr-st"
+      id: "facebook/seamless-m4t-v2-large"
     }
   ],
   spaces: [
@@ -2954,8 +2954,8 @@ var taskData24 = {
   canonicalId: "text-to-audio",
   datasets: [
     {
-      description: "Thousands of short audio clips of a single speaker.",
-      id: "lj_speech"
+      description: "10K hours of multi-speaker English dataset.",
+      id: "parler-tts/mls_eng_10k"
     },
     {
       description: "Multi-speaker English dataset.",
@@ -2993,8 +2993,8 @@ var taskData24 = {
       id: "facebook/mms-tts"
     },
     {
-      description: "An end-to-end speech synthesis model.",
-      id: "microsoft/speecht5_tts"
+      description: "A prompt based, powerful TTS model.",
+      id: "parler-tts/parler_tts_mini_v0.1"
     }
   ],
   spaces: [
@@ -3007,8 +3007,8 @@ var taskData24 = {
       id: "coqui/xtts"
     },
     {
-      description: "An application that synthesizes speech for various speaker types.",
-      id: "Matthijs/speecht5-tts-demo"
+      description: "An application that synthesizes speech for diverse speaker prompts.",
+      id: "parler-tts/parler_tts_mini"
     }
   ],
   summary: "Text-to-Speech (TTS) is the task of generating natural sounding speech given text input. TTS models can be extended to have a single model that generates speech for multiple speakers and multiple languages.",
@@ -5438,6 +5438,30 @@ var SKUS = {
         tflops: 91.1,
         memory: [48]
       },
+      "RTX 5880 Ada": {
+        tflops: 69.3,
+        memory: [48]
+      },
+      "RTX 5000 Ada": {
+        tflops: 65.3,
+        memory: [32]
+      },
+      "RTX 4500 Ada": {
+        tflops: 39.6,
+        memory: [24]
+      },
+      "RTX 4000 Ada": {
+        tflops: 26.7,
+        memory: [20]
+      },
+      "RTX 4000 SFF Ada": {
+        tflops: 19.2,
+        memory: [20]
+      },
+      "RTX 2000 Ada": {
+        tflops: 12,
+        memory: [16]
+      },
       A100: {
         tflops: 77.97,
         memory: [80, 40]
@@ -5450,14 +5474,14 @@ var SKUS = {
         tflops: 31.24,
         memory: [24]
       },
-      T4: {
-        tflops: 65.13,
-        memory: [16]
-      },
       "RTX 4090": {
         tflops: 82.58,
         memory: [24]
       },
+      "RTX 4090D": {
+        tflops: 79.49,
+        memory: [24]
+      },
       "RTX 4080 SUPER": {
         tflops: 52.2,
         memory: [16]
@@ -5482,6 +5506,14 @@ var SKUS = {
         tflops: 44.1,
         memory: [16]
       },
+      "RTX 4060": {
+        tflops: 15.11,
+        memory: [8]
+      },
+      "RTX 4060 Ti": {
+        tflops: 22.06,
+        memory: [8, 16]
+      },
       "RTX 3090": {
         tflops: 35.58,
         memory: [24]
@@ -5518,13 +5550,53 @@ var SKUS = {
         tflops: 12.74,
         memory: [12, 8]
       },
+      "RTX 2070": {
+        tflops: 14.93,
+        memory: [8]
+      },
+      "RTX 3050 Mobile": {
+        tflops: 7.639,
+        memory: [6]
+      },
       "RTX 2060 Mobile": {
         tflops: 9.22,
         memory: [6]
       },
+      "GTX 1080 Ti": {
+        tflops: 11.34,
+        // float32 (GPU does not support native float16)
+        memory: [11]
+      },
+      "GTX 1070 Ti": {
+        tflops: 8.2,
+        // float32 (GPU does not support native float16)
+        memory: [8]
+      },
       "RTX Titan": {
         tflops: 32.62,
         memory: [24]
+      },
+      "GTX 1650 Mobile": {
+        tflops: 6.39,
+        memory: [4]
+      },
+      T4: {
+        tflops: 65.13,
+        memory: [16]
+      },
+      V100: {
+        tflops: 28.26,
+        memory: [32, 16]
+      },
+      "Quadro P6000": {
+        tflops: 12.63,
+        // float32 (GPU does not support native float16)
+        memory: [24]
+      },
+      P40: {
+        tflops: 11.76,
+        // float32 (GPU does not support native float16)
+        memory: [24]
       }
     },
     AMD: {
@@ -5540,6 +5612,10 @@ var SKUS = {
         tflops: 181,
         memory: [64]
       },
+      MI100: {
+        tflops: 184.6,
+        memory: [32]
+      },
       "RX 7900 XTX": {
         tflops: 122.8,
         memory: [24]
@@ -5563,6 +5639,18 @@ var SKUS = {
       "RX 7600 XT": {
         tflops: 45.14,
         memory: [16, 8]
+      },
+      "RX 6950 XT": {
+        tflops: 47.31,
+        memory: [16]
+      },
+      "RX 6800": {
+        tflops: 32.33,
+        memory: [16]
+      },
+      "Radeon Pro VII": {
+        tflops: 26.11,
+        memory: [16]
       }
     }
   },
@@ -5773,6 +5861,13 @@ var LOCAL_APPS = {
     displayOnModelPage: isGgufModel,
     deeplink: (model) => new URL(`https://backyard.ai/hf/model/${model.id}`)
   },
+  sanctum: {
+    prettyLabel: "Sanctum",
+    docsUrl: "https://sanctum.ai",
+    mainTask: "text-generation",
+    displayOnModelPage: isGgufModel,
+    deeplink: (model) => new URL(`sanctum://open_from_hf?model=${model.id}`)
+  },
   drawthings: {
     prettyLabel: "Draw Things",
     docsUrl: "https://drawthings.ai",

package/dist/src/hardware.d.ts

@@ -44,6 +44,30 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
+            "RTX 5880 Ada": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 5000 Ada": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 4500 Ada": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 4000 Ada": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 4000 SFF Ada": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 2000 Ada": {
+                tflops: number;
+                memory: number[];
+            };
             A100: {
                 tflops: number;
                 memory: number[];
@@ -56,11 +80,11 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
-            T4: {
+            "RTX 4090": {
                 tflops: number;
                 memory: number[];
             };
-            "RTX 4090": {
+            "RTX 4090D": {
                 tflops: number;
                 memory: number[];
             };
@@ -88,6 +112,14 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
+            "RTX 4060": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 4060 Ti": {
+                tflops: number;
+                memory: number[];
+            };
             "RTX 3090": {
                 tflops: number;
                 memory: number[];
@@ -124,14 +156,50 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
+            "RTX 2070": {
+                tflops: number;
+                memory: number[];
+            };
+            "RTX 3050 Mobile": {
+                tflops: number;
+                memory: number[];
+            };
             "RTX 2060 Mobile": {
                 tflops: number;
                 memory: number[];
             };
+            "GTX 1080 Ti": {
+                tflops: number;
+                memory: number[];
+            };
+            "GTX 1070 Ti": {
+                tflops: number;
+                memory: number[];
+            };
             "RTX Titan": {
                 tflops: number;
                 memory: number[];
             };
+            "GTX 1650 Mobile": {
+                tflops: number;
+                memory: number[];
+            };
+            T4: {
+                tflops: number;
+                memory: number[];
+            };
+            V100: {
+                tflops: number;
+                memory: number[];
+            };
+            "Quadro P6000": {
+                tflops: number;
+                memory: number[];
+            };
+            P40: {
+                tflops: number;
+                memory: number[];
+            };
         };
         AMD: {
             MI300: {
@@ -146,6 +214,10 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
+            MI100: {
+                tflops: number;
+                memory: number[];
+            };
             "RX 7900 XTX": {
                 tflops: number;
                 memory: number[];
@@ -170,6 +242,18 @@ export declare const SKUS: {
                 tflops: number;
                 memory: number[];
             };
+            "RX 6950 XT": {
+                tflops: number;
+                memory: number[];
+            };
+            "RX 6800": {
+                tflops: number;
+                memory: number[];
+            };
+            "Radeon Pro VII": {
+                tflops: number;
+                memory: number[];
+            };
         };
     };
     CPU: {

package/dist/src/hardware.d.ts.map

@@ -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;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA2SuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}
+{"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;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CA+XuD,CAAC;AAEzE,MAAM,MAAM,OAAO,GAAG,MAAM,OAAO,IAAI,CAAC"}

package/dist/src/local-apps.d.ts

@@ -77,6 +77,13 @@ export declare const LOCAL_APPS: {
         displayOnModelPage: typeof isGgufModel;
         deeplink: (model: ModelData) => URL;
     };
+    sanctum: {
+        prettyLabel: string;
+        docsUrl: string;
+        mainTask: "text-generation";
+        displayOnModelPage: typeof isGgufModel;
+        deeplink: (model: ModelData) => URL;
+    };
     drawthings: {
         prettyLabel: string;
         docsUrl: string;

package/dist/src/local-apps.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"local-apps.d.ts","sourceRoot":"","sources":["../../src/local-apps.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,cAAc,CAAC;AAC9C,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAEhD;;GAEG;AACH,MAAM,MAAM,QAAQ,GAAG;IACtB;;OAEG;IACH,WAAW,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,OAAO,EAAE,MAAM,CAAC;IAChB;;OAEG;IACH,QAAQ,EAAE,YAAY,CAAC;IACvB;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IAEpB,UAAU,CAAC,EAAE,OAAO,CAAC;IACrB;;OAEG;IACH,kBAAkB,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,OAAO,CAAC;CAClD,GAAG,CACD;IACA;;OAEG;IACH,QAAQ,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,GAAG,CAAC;CACnC,GACD;IACA;;OAEG;IACH,OAAO,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,MAAM,GAAG,MAAM,EAAE,CAAC;CAChD,CACH,CAAC;AAEF,iBAAS,WAAW,CAAC,KAAK,EAAE,SAAS,WAEpC;AAmBD;;;;;;;;;;GAUG;AACH,eAAO,MAAM,UAAU;;;;;;yBA5BS,SAAS,KAAG,MAAM,EAAE;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAiFhB,CAAC;AAErC,MAAM,MAAM,WAAW,GAAG,MAAM,OAAO,UAAU,CAAC"}
+{"version":3,"file":"local-apps.d.ts","sourceRoot":"","sources":["../../src/local-apps.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,SAAS,EAAE,MAAM,cAAc,CAAC;AAC9C,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,aAAa,CAAC;AAEhD;;GAEG;AACH,MAAM,MAAM,QAAQ,GAAG;IACtB;;OAEG;IACH,WAAW,EAAE,MAAM,CAAC;IACpB;;OAEG;IACH,OAAO,EAAE,MAAM,CAAC;IAChB;;OAEG;IACH,QAAQ,EAAE,YAAY,CAAC;IACvB;;OAEG;IACH,SAAS,CAAC,EAAE,OAAO,CAAC;IAEpB,UAAU,CAAC,EAAE,OAAO,CAAC;IACrB;;OAEG;IACH,kBAAkB,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,OAAO,CAAC;CAClD,GAAG,CACD;IACA;;OAEG;IACH,QAAQ,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,GAAG,CAAC;CACnC,GACD;IACA;;OAEG;IACH,OAAO,EAAE,CAAC,KAAK,EAAE,SAAS,KAAK,MAAM,GAAG,MAAM,EAAE,CAAC;CAChD,CACH,CAAC;AAEF,iBAAS,WAAW,CAAC,KAAK,EAAE,SAAS,WAEpC;AAmBD;;;;;;;;;;GAUG;AACH,eAAO,MAAM,UAAU;;;;;;yBA5BS,SAAS,KAAG,MAAM,EAAE;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;CAwFhB,CAAC;AAErC,MAAM,MAAM,WAAW,GAAG,MAAM,OAAO,UAAU,CAAC"}

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@huggingface/tasks",
   "packageManager": "pnpm@8.10.5",
-  "version": "0.10.8",
+  "version": "0.10.9",
   "description": "List of ML tasks for huggingface.co/tasks",
   "repository": "https://github.com/huggingface/huggingface.js.git",
   "publishConfig": {

package/src/hardware.ts

@@ -48,6 +48,30 @@ export const SKUS = {
 				tflops: 91.1,
 				memory: [48],
 			},
+			"RTX 5880 Ada": {
+				tflops: 69.3,
+				memory: [48],
+			},
+			"RTX 5000 Ada": {
+				tflops: 65.3,
+				memory: [32],
+			},
+			"RTX 4500 Ada": {
+				tflops: 39.6,
+				memory: [24],
+			},
+			"RTX 4000 Ada": {
+				tflops: 26.7,
+				memory: [20],
+			},
+			"RTX 4000 SFF Ada": {
+				tflops: 19.2,
+				memory: [20],
+			},
+			"RTX 2000 Ada": {
+				tflops: 12.0,
+				memory: [16],
+			},
 			A100: {
 				tflops: 77.97,
 				memory: [80, 40],
@@ -60,14 +84,14 @@ export const SKUS = {
 				tflops: 31.24,
 				memory: [24],
 			},
-			T4: {
-				tflops: 65.13,
-				memory: [16],
-			},
 			"RTX 4090": {
 				tflops: 82.58,
 				memory: [24],
 			},
+			"RTX 4090D": {
+				tflops: 79.49,
+				memory: [24],
+			},
 			"RTX 4080 SUPER": {
 				tflops: 52.2,
 				memory: [16],
@@ -92,6 +116,14 @@ export const SKUS = {
 				tflops: 44.1,
 				memory: [16],
 			},
+			"RTX 4060": {
+				tflops: 15.11,
+				memory: [8],
+			},
+			"RTX 4060 Ti": {
+				tflops: 22.06,
+				memory: [8, 16],
+			},
 			"RTX 3090": {
 				tflops: 35.58,
 				memory: [24],
@@ -128,14 +160,50 @@ export const SKUS = {
 				tflops: 12.74,
 				memory: [12, 8],
 			},
+			"RTX 2070": {
+				tflops: 14.93,
+				memory: [8],
+			},
+			"RTX 3050 Mobile": {
+				tflops: 7.639,
+				memory: [6],
+			},
 			"RTX 2060 Mobile": {
 				tflops: 9.22,
 				memory: [6],
 			},
+			"GTX 1080 Ti": {
+				tflops: 11.34, // float32 (GPU does not support native float16)
+				memory: [11],
+			},
+			"GTX 1070 Ti": {
+				tflops: 8.2, // float32 (GPU does not support native float16)
+				memory: [8],
+			},
 			"RTX Titan": {
 				tflops: 32.62,
 				memory: [24],
 			},
+			"GTX 1650 Mobile": {
+				tflops: 6.39,
+				memory: [4],
+			},
+			T4: {
+				tflops: 65.13,
+				memory: [16],
+			},
+			V100: {
+				tflops: 28.26,
+				memory: [32, 16],
+			},
+			"Quadro P6000": {
+				tflops: 12.63, // float32 (GPU does not support native float16)
+				memory: [24],
+			},
+			P40: {
+				tflops: 11.76, // float32 (GPU does not support native float16)
+				memory: [24],
+			},
 		},
 		AMD: {
 			MI300: {
@@ -150,6 +218,10 @@ export const SKUS = {
 				tflops: 181.0,
 				memory: [64],
 			},
+			MI100: {
+				tflops: 184.6,
+				memory: [32],
+			},
 			"RX 7900 XTX": {
 				tflops: 122.8,
 				memory: [24],
@@ -174,6 +246,18 @@ export const SKUS = {
 				tflops: 45.14,
 				memory: [16, 8],
 			},
+			"RX 6950 XT": {
+				tflops: 47.31,
+				memory: [16],
+			},
+			"RX 6800": {
+				tflops: 32.33,
+				memory: [16],
+			},
+			"Radeon Pro VII": {
+				tflops: 26.11,
+				memory: [16],
+			},
 		},
 	},
 	CPU: {

package/src/local-apps.ts

@@ -103,6 +103,13 @@ export const LOCAL_APPS = {
 		displayOnModelPage: isGgufModel,
 		deeplink: (model) => new URL(`https://backyard.ai/hf/model/${model.id}`),
 	},
+	sanctum: {
+		prettyLabel: "Sanctum",
+		docsUrl: "https://sanctum.ai",
+		mainTask: "text-generation",
+		displayOnModelPage: isGgufModel,
+		deeplink: (model) => new URL(`sanctum://open_from_hf?model=${model.id}`),
+	},
 	drawthings: {
 		prettyLabel: "Draw Things",
 		docsUrl: "https://drawthings.ai",

package/src/tasks/automatic-speech-recognition/about.md

@@ -18,7 +18,7 @@ The use of Multilingual ASR has become popular, the idea of maintaining just a s
 
 ## Inference
 
-The Hub contains over [~9,000 ASR models](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=downloads) that you can use right away by trying out the widgets directly in the browser or calling the models as a service using Inference Endpoints. Here is a simple code snippet to do exactly this:
+The Hub contains over [17,000 ASR models](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=downloads) that you can test right away in your browser using the model page widgets. You can also use any model as a service using the Serverless Inference API. We also support libraries such as [transformers](https://huggingface.co/models?library=transformers&pipeline_tag=automatic-speech-recognition&sort=downloads), [speechbrain](https://huggingface.co/models?library=speechbrain&pipeline_tag=automatic-speech-recognition&sort=downloads), [NeMo](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&library=nemo&sort=downloads) and [espnet](https://huggingface.co/models?library=espnet&pipeline_tag=automatic-speech-recognition&sort=downloads) via the Serverless Inference API. Here's a simple code snippet to run inference:
 
 ```python
 import json
@@ -36,20 +36,7 @@ def query(filename):
 data = query("sample1.flac")
 ```
 
-You can also use libraries such as [transformers](https://huggingface.co/models?library=transformers&pipeline_tag=automatic-speech-recognition&sort=downloads), [speechbrain](https://huggingface.co/models?library=speechbrain&pipeline_tag=automatic-speech-recognition&sort=downloads), [NeMo](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&library=nemo&sort=downloads) and [espnet](https://huggingface.co/models?library=espnet&pipeline_tag=automatic-speech-recognition&sort=downloads) if you want one-click managed Inference without any hassle.
-
-```python
-from transformers import pipeline
-
-with open("sample.flac", "rb") as f:
-  data = f.read()
-
-pipe = pipeline("automatic-speech-recognition", "openai/whisper-large-v2")
-pipe("sample.flac")
-# {'text': "GOING ALONG SLUSHY COUNTRY ROADS AND SPEAKING TO DAMP AUDIENCES IN DRAUGHTY SCHOOL ROOMS DAY AFTER DAY FOR A FORTNIGHT HE'LL HAVE TO PUT IN AN APPEARANCE AT SOME PLACE OF WORSHIP ON SUNDAY MORNING AND HE CAN COME TO US IMMEDIATELY AFTERWARDS"}
-```
-
-You can use [huggingface.js](https://github.com/huggingface/huggingface.js) to transcribe text with javascript using models on Hugging Face Hub.
+You can also use[huggingface.js](https://github.com/huggingface/huggingface.js), the JavaScript client, to transcribe models with the Inference API.
 
 ```javascript
 import { HfInference } from "@huggingface/inference";
@@ -57,10 +44,23 @@ import { HfInference } from "@huggingface/inference";
 const inference = new HfInference(HF_TOKEN);
 await inference.automaticSpeechRecognition({
 	data: await (await fetch("sample.flac")).blob(),
-	model: "openai/whisper-large-v2",
+	model: "openai/whisper-large-v3",
 });
 ```
 
+For transformers compatible models like Whisper, Wav2Vec2, HuBERT, etc. You can also run inference in Python using transformers as follows:
+
+```python
+# pip install --upgrade transformers
+
+from transformers import pipeline
+
+pipe = pipeline("automatic-speech-recognition", "openai/whisper-large-v3")
+
+pipe("sample.flac")
+# {'text': "GOING ALONG SLUSHY COUNTRY ROADS AND SPEAKING TO DAMP AUDIENCES IN DRAUGHTY SCHOOL ROOMS DAY AFTER DAY FOR A FORTNIGHT HE'LL HAVE TO PUT IN AN APPEARANCE AT SOME PLACE OF WORSHIP ON SUNDAY MORNING AND HE CAN COME TO US IMMEDIATELY AFTERWARDS"}
+```
+
 ## Solving ASR for your own data
 
 We have some great news! You can fine-tune (transfer learning) a foundational speech model on a specific language without tonnes of data. Pretrained models such as Whisper, Wav2Vec2-MMS and HuBERT exist. [OpenAI's Whisper model](https://huggingface.co/openai/whisper-large-v3) is a large multilingual model trained on 100+ languages and with 4 Million hours of speech.

package/src/tasks/automatic-speech-recognition/data.ts

@@ -3,16 +3,16 @@ import type { TaskDataCustom } from "..";
 const taskData: TaskDataCustom = {
 	datasets: [
 		{
-			description: "18,000 hours of multilingual audio-text dataset in 108 languages.",
-			id: "mozilla-foundation/common_voice_13_0",
+			description: "31,175 hours of multilingual audio-text dataset in 108 languages.",
+			id: "mozilla-foundation/common_voice_17_0",
 		},
 		{
 			description: "An English dataset with 1,000 hours of data.",
 			id: "librispeech_asr",
 		},
 		{
-			description: "High quality, multi-speaker audio data and their transcriptions in various languages.",
-			id: "openslr",
+			description: "A multi-lingual audio dataset with 370K hours of audio.",
+			id: "espnet/yodas",
 		},
 	],
 	demo: {
@@ -47,12 +47,12 @@ const taskData: TaskDataCustom = {
 			id: "openai/whisper-large-v3",
 		},
 		{
-			description: "A good generic ASR model by MetaAI.",
-			id: "facebook/wav2vec2-base-960h",
+			description: "A good generic speech model by MetaAI for fine-tuning.",
+			id: "facebook/w2v-bert-2.0",
 		},
 		{
 			description: "An end-to-end model that performs ASR and Speech Translation by MetaAI.",
-			id: "facebook/s2t-small-mustc-en-fr-st",
+			id: "facebook/seamless-m4t-v2-large",
 		},
 	],
 	spaces: [

package/src/tasks/text-to-speech/about.md

@@ -58,8 +58,6 @@ await inference.textToSpeech({
 
 - [Hugging Face Audio Course](https://huggingface.co/learn/audio-course/chapter6/introduction)
 - [ML for Audio Study Group - Text to Speech Deep Dive](https://www.youtube.com/watch?v=aLBedWj-5CQ)
-- [An introduction to SpeechT5, a multi-purpose speech recognition and synthesis model](https://huggingface.co/blog/speecht5).
-- [A guide on Fine-tuning Whisper For Multilingual ASR with 🤗Transformers](https://huggingface.co/blog/fine-tune-whisper)
 - [Speech Synthesis, Recognition, and More With SpeechT5](https://huggingface.co/blog/speecht5)
 - [Optimizing a Text-To-Speech model using 🤗 Transformers](https://huggingface.co/blog/optimizing-bark)
--
+- [Train your own TTS models with Parler-TTS](https://github.com/huggingface/parler-tts)

package/src/tasks/text-to-speech/data.ts

@@ -4,8 +4,8 @@ const taskData: TaskDataCustom = {
 	canonicalId: "text-to-audio",
 	datasets: [
 		{
-			description: "Thousands of short audio clips of a single speaker.",
-			id: "lj_speech",
+			description: "10K hours of multi-speaker English dataset.",
+			id: "parler-tts/mls_eng_10k",
 		},
 		{
 			description: "Multi-speaker English dataset.",
@@ -43,8 +43,8 @@ const taskData: TaskDataCustom = {
 			id: "facebook/mms-tts",
 		},
 		{
-			description: "An end-to-end speech synthesis model.",
-			id: "microsoft/speecht5_tts",
+			description: "A prompt based, powerful TTS model.",
+			id: "parler-tts/parler_tts_mini_v0.1",
 		},
 	],
 	spaces: [
@@ -57,8 +57,8 @@ const taskData: TaskDataCustom = {
 			id: "coqui/xtts",
 		},
 		{
-			description: "An application that synthesizes speech for various speaker types.",
-			id: "Matthijs/speecht5-tts-demo",
+			description: "An application that synthesizes speech for diverse speaker prompts.",
+			id: "parler-tts/parler_tts_mini",
 		},
 	],
 	summary: