@huggingface/transformers 3.2.3 → 3.3.0

package/dist/transformers.cjs

@@ -4158,23 +4158,23 @@ class FeatureExtractor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Ca
     }
 
     /**
-     * Instantiate one of the processor classes of the library from a pretrained model.
+     * Instantiate one of the feature extractor classes of the library from a pretrained model.
      * 
-     * The processor class to instantiate is selected based on the `image_processor_type` (or `feature_extractor_type`; legacy)
-     * property of the config object (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
+     * The feature extractor class to instantiate is selected based on the `feature_extractor_type` property of
+     * the config object (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
      * 
      * @param {string} pretrained_model_name_or_path The name or path of the pretrained model. Can be either:
-     * - A string, the *model id* of a pretrained processor hosted inside a model repo on huggingface.co.
+     * - A string, the *model id* of a pretrained feature_extractor hosted inside a model repo on huggingface.co.
      *   Valid model ids can be located at the root-level, like `bert-base-uncased`, or namespaced under a
      *   user or organization name, like `dbmdz/bert-base-german-cased`.
-     * - A path to a *directory* containing processor files, e.g., `./my_model_directory/`.
-     * @param {import('../utils/hub.js').PretrainedOptions} options Additional options for loading the processor.
+     * - A path to a *directory* containing feature_extractor files, e.g., `./my_model_directory/`.
+     * @param {import('../utils/hub.js').PretrainedOptions} options Additional options for loading the feature_extractor.
      * 
-     * @returns {Promise<FeatureExtractor>} A new instance of the Processor class.
+     * @returns {Promise<FeatureExtractor>} A new instance of the Feature Extractor class.
      */
     static async from_pretrained(pretrained_model_name_or_path, options) {
-        const preprocessorConfig = await (0,_utils_hub_js__WEBPACK_IMPORTED_MODULE_2__.getModelJSON)(pretrained_model_name_or_path, _utils_constants_js__WEBPACK_IMPORTED_MODULE_0__.FEATURE_EXTRACTOR_NAME, true, options);
-        return new this(preprocessorConfig);
+        const config = await (0,_utils_hub_js__WEBPACK_IMPORTED_MODULE_2__.getModelJSON)(pretrained_model_name_or_path, _utils_constants_js__WEBPACK_IMPORTED_MODULE_0__.FEATURE_EXTRACTOR_NAME, true, options);
+        return new this(config);
     }
 }
 
@@ -4207,6 +4207,7 @@ function validate_audio_inputs(audio, feature_extractor) {
 __webpack_require__.r(__webpack_exports__);
 /* harmony export */ __webpack_require__.d(__webpack_exports__, {
 /* harmony export */   ImageProcessor: () => (/* binding */ ImageProcessor),
+/* harmony export */   center_to_corners_format: () => (/* binding */ center_to_corners_format),
 /* harmony export */   post_process_instance_segmentation: () => (/* binding */ post_process_instance_segmentation),
 /* harmony export */   post_process_object_detection: () => (/* binding */ post_process_object_detection),
 /* harmony export */   post_process_panoptic_segmentation: () => (/* binding */ post_process_panoptic_segmentation),
@@ -4825,14 +4826,20 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
         this.do_thumbnail = config.do_thumbnail;
         this.size = config.size ?? config.image_size;
         this.do_resize = config.do_resize ?? (this.size !== undefined);
+        // @ts-expect-error TS2339
         this.size_divisibility = config.size_divisibility ?? config.size_divisor;
 
         this.do_center_crop = config.do_center_crop;
+        // @ts-expect-error TS2339
         this.crop_size = config.crop_size;
+        // @ts-expect-error TS2339
         this.do_convert_rgb = config.do_convert_rgb ?? true;
+        // @ts-expect-error TS2339
         this.do_crop_margin = config.do_crop_margin;
 
+        // @ts-expect-error TS2339
         this.pad_size = config.pad_size;
+        // @ts-expect-error TS2339
         this.do_pad = config.do_pad;
 
         if (this.do_pad && !this.pad_size && this.size && this.size.width !== undefined && this.size.height !== undefined) {
@@ -5041,6 +5048,7 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
         // Support both formats for backwards compatibility
         else if (Number.isInteger(size)) {
             shortest_edge = size;
+            // @ts-expect-error TS2339
             longest_edge = this.config.max_size ?? shortest_edge;
 
         } else if (size !== undefined) {
@@ -5109,6 +5117,7 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
         } else if (size.min_pixels !== undefined && size.max_pixels !== undefined) {
             // Custom resize logic for Qwen2-VL models
             const { min_pixels, max_pixels } = size;
+            // @ts-expect-error TS2339
             const factor = this.config.patch_size * this.config.merge_size;
             return smart_resize(srcHeight, srcWidth, factor, min_pixels, max_pixels);
         } else {
@@ -5124,6 +5133,7 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
     async resize(image) {
         const [newWidth, newHeight] = this.get_resize_output_image_size(image, this.size);
         return await image.resize(newWidth, newHeight, {
+            // @ts-expect-error TS2322
             resample: this.resample,
         });
     }
@@ -5174,6 +5184,7 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
 
         // Resize the image using thumbnail method.
         if (this.do_thumbnail) {
+            // @ts-expect-error TS2345
             image = await this.thumbnail(image, this.size, this.resample);
         }
 
@@ -5198,6 +5209,7 @@ class ImageProcessor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__.Call
         // NOTE: All pixel-level manipulation (i.e., modifying `pixelData`)
         // occurs with data in the hwc format (height, width, channels), 
         // to emulate the behavior of the original Python code (w/ numpy).
+        /** @type {Float32Array} */
         let pixelData = Float32Array.from(image.data);
         let imgDims = [image.height, image.width, image.channels];
 
@@ -5356,6 +5368,7 @@ __webpack_require__.r(__webpack_exports__);
 /**
  * @typedef {Object} ProcessorProperties Additional processor-specific properties.
  * @typedef {import('../utils/hub.js').PretrainedOptions & ProcessorProperties} PretrainedProcessorOptions
+ * @typedef {import('../tokenizers.js').PreTrainedTokenizer} PreTrainedTokenizer
  */
 
 
@@ -5389,7 +5402,7 @@ class Processor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Callable
     }
 
     /**
-     * @returns {import('../tokenizers.js').PreTrainedTokenizer|undefined} The tokenizer of the processor, if it exists.
+     * @returns {PreTrainedTokenizer|undefined} The tokenizer of the processor, if it exists.
      */
     get tokenizer() {
         return this.components.tokenizer;
@@ -5402,6 +5415,11 @@ class Processor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Callable
         return this.components.feature_extractor;
     }
 
+    /**
+     * @param {Parameters<PreTrainedTokenizer['apply_chat_template']>[0]} messages
+     * @param {Parameters<PreTrainedTokenizer['apply_chat_template']>[1]} options
+     * @returns {ReturnType<PreTrainedTokenizer['apply_chat_template']>}
+     */
     apply_chat_template(messages, options = {}) {
         if (!this.tokenizer) {
             throw new Error('Unable to apply chat template without a tokenizer.');
@@ -5412,6 +5430,10 @@ class Processor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Callable
         });
     }
 
+    /**
+     * @param {Parameters<PreTrainedTokenizer['batch_decode']>} args
+     * @returns {ReturnType<PreTrainedTokenizer['batch_decode']>}
+     */
     batch_decode(...args) {
         if (!this.tokenizer) {
             throw new Error('Unable to decode without a tokenizer.');
@@ -5419,6 +5441,17 @@ class Processor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Callable
         return this.tokenizer.batch_decode(...args);
     }
 
+    /**
+     * @param {Parameters<PreTrainedTokenizer['decode']>} args
+     * @returns {ReturnType<PreTrainedTokenizer['decode']>}
+     */
+    decode(...args) {
+        if (!this.tokenizer) {
+            throw new Error('Unable to decode without a tokenizer.');
+        }
+        return this.tokenizer.decode(...args);
+    }
+
 
     /**
      * Calls the feature_extractor function with the given input.
@@ -5439,8 +5472,8 @@ class Processor extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_1__.Callable
     /**
      * Instantiate one of the processor classes of the library from a pretrained model.
      * 
-     * The processor class to instantiate is selected based on the `feature_extractor_type` property of the config object
-     * (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
+     * The processor class to instantiate is selected based on the `image_processor_type` (or `feature_extractor_type`; legacy)
+     * property of the config object (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
      * 
      * @param {string} pretrained_model_name_or_path The name or path of the pretrained model. Can be either:
      * - A string, the *model id* of a pretrained processor hosted inside a model repo on huggingface.co.
@@ -5560,15 +5593,19 @@ function getNormalizedConfig(config) {
         case 'florence2':
         case 'llava_onevision':
         case 'idefics3':
+            // @ts-expect-error TS2339
             init_normalized_config = getNormalizedConfig(config.text_config);
             break;
         case 'moondream1':
+            // @ts-expect-error TS2339
             init_normalized_config = getNormalizedConfig(config.phi_config);
             break;
         case 'musicgen':
+            // @ts-expect-error TS2339
             init_normalized_config = getNormalizedConfig(config.decoder);
             break;
         case 'multi_modality':
+            // @ts-expect-error TS2339
             init_normalized_config = getNormalizedConfig(config.language_config);
             break;
 
@@ -5689,6 +5726,7 @@ function getNormalizedConfig(config) {
             break;
 
         case 'vision-encoder-decoder':
+            // @ts-expect-error TS2339
             const decoderConfig = getNormalizedConfig(config.decoder);
 
             const add_encoder_pkv = 'num_decoder_layers' in decoderConfig;
@@ -5932,7 +5970,7 @@ __webpack_require__.r(__webpack_exports__);
 
 
 
-const VERSION = '3.2.3';
+const VERSION = '3.3.0';
 
 // Check if various APIs are available (depends on environment)
 const IS_BROWSER_ENV = typeof window !== "undefined" && typeof window.document !== "undefined";
@@ -6068,7 +6106,6 @@ function isEmpty(obj) {
 }
 
 
-
 /***/ }),
 
 /***/ "./src/generation/configuration_utils.js":
@@ -7683,6 +7720,7 @@ class TextStreamer extends BaseStreamer {
      * @param {import('../tokenizers.js').PreTrainedTokenizer} tokenizer
      * @param {Object} options
      * @param {boolean} [options.skip_prompt=false] Whether to skip the prompt tokens
+     * @param {boolean} [options.skip_special_tokens=true] Whether to skip special tokens when decoding
      * @param {function(string): void} [options.callback_function=null] Function to call when a piece of text is ready to display
      * @param {function(bigint[]): void} [options.token_callback_function=null] Function to call when a new token is generated
      * @param {Object} [options.decode_kwargs={}] Additional keyword arguments to pass to the tokenizer's decode method
@@ -7691,6 +7729,7 @@ class TextStreamer extends BaseStreamer {
         skip_prompt = false,
         callback_function = null,
         token_callback_function = null,
+        skip_special_tokens = true,
         decode_kwargs = {},
         ...kwargs
     } = {}) {
@@ -7699,7 +7738,7 @@ class TextStreamer extends BaseStreamer {
         this.skip_prompt = skip_prompt;
         this.callback_function = callback_function ?? stdout_write;
         this.token_callback_function = token_callback_function;
-        this.decode_kwargs = { ...decode_kwargs, ...kwargs };
+        this.decode_kwargs = { skip_special_tokens, ...decode_kwargs, ...kwargs };
 
         // variables used in the streaming process
         this.token_cache = [];
@@ -7815,9 +7854,10 @@ class WhisperTextStreamer extends TextStreamer {
     } = {}) {
         super(tokenizer, {
             skip_prompt,
+            skip_special_tokens,
             callback_function,
             token_callback_function,
-            decode_kwargs: { skip_special_tokens, ...decode_kwargs },
+            decode_kwargs,
         });
         this.timestamp_begin = tokenizer.timestamp_begin;
 
@@ -8071,6 +8111,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   GraniteForCausalLM: () => (/* binding */ GraniteForCausalLM),
 /* harmony export */   GraniteModel: () => (/* binding */ GraniteModel),
 /* harmony export */   GranitePreTrainedModel: () => (/* binding */ GranitePreTrainedModel),
+/* harmony export */   GroundingDinoForObjectDetection: () => (/* binding */ GroundingDinoForObjectDetection),
+/* harmony export */   GroundingDinoPreTrainedModel: () => (/* binding */ GroundingDinoPreTrainedModel),
 /* harmony export */   GroupViTModel: () => (/* binding */ GroupViTModel),
 /* harmony export */   GroupViTPreTrainedModel: () => (/* binding */ GroupViTPreTrainedModel),
 /* harmony export */   HieraForImageClassification: () => (/* binding */ HieraForImageClassification),
@@ -8279,6 +8321,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   Starcoder2ForCausalLM: () => (/* binding */ Starcoder2ForCausalLM),
 /* harmony export */   Starcoder2Model: () => (/* binding */ Starcoder2Model),
 /* harmony export */   Starcoder2PreTrainedModel: () => (/* binding */ Starcoder2PreTrainedModel),
+/* harmony export */   StyleTextToSpeech2Model: () => (/* binding */ StyleTextToSpeech2Model),
+/* harmony export */   StyleTextToSpeech2PreTrainedModel: () => (/* binding */ StyleTextToSpeech2PreTrainedModel),
 /* harmony export */   Swin2SRForImageSuperResolution: () => (/* binding */ Swin2SRForImageSuperResolution),
 /* harmony export */   Swin2SRModel: () => (/* binding */ Swin2SRModel),
 /* harmony export */   Swin2SRPreTrainedModel: () => (/* binding */ Swin2SRPreTrainedModel),
@@ -8594,8 +8638,11 @@ async function getSession(pretrained_model_name_or_path, fileName, options) {
     } else if (session_options.externalData !== undefined) {
         externalDataPromises = session_options.externalData.map(async (ext) => {
             // if the external data is a string, fetch the file and replace the string with its content
+            // @ts-expect-error TS2339
             if (typeof ext.data === "string") {
+                // @ts-expect-error TS2339
                 const ext_buffer = await (0,_utils_hub_js__WEBPACK_IMPORTED_MODULE_5__.getModelFile)(pretrained_model_name_or_path, ext.data, true, options);
+                // @ts-expect-error TS2698
                 return { ...ext, data: ext_buffer };
             }
             return ext;
@@ -8853,14 +8900,23 @@ async function encoderForward(self, model_inputs) {
         encoderFeeds.inputs_embeds = await self.encode_text({ input_ids: model_inputs.input_ids });
     }
     if (session.inputNames.includes('token_type_ids') && !encoderFeeds.token_type_ids) {
+        if (!encoderFeeds.input_ids) {
+            throw new Error('Both `input_ids` and `token_type_ids` are missing in the model inputs.');
+        }
         // Assign default `token_type_ids` (all zeroes) to the `encoderFeeds` if the model expects it,
         // but they weren't created by the tokenizer.
-        encoderFeeds.token_type_ids = new _utils_tensor_js__WEBPACK_IMPORTED_MODULE_9__.Tensor(
-            'int64',
-            new BigInt64Array(encoderFeeds.input_ids.data.length),
-            encoderFeeds.input_ids.dims
-        )
+        encoderFeeds.token_type_ids = (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_9__.zeros_like)(encoderFeeds.input_ids);
     }
+    if (session.inputNames.includes('pixel_mask') && !encoderFeeds.pixel_mask) {
+        if (!encoderFeeds.pixel_values) {
+            throw new Error('Both `pixel_values` and `pixel_mask` are missing in the model inputs.');
+        }
+        // Assign default `pixel_mask` (all ones) to the `encoderFeeds` if the model expects it,
+        // but they weren't created by the processor.
+        const dims = encoderFeeds.pixel_values.dims;
+        encoderFeeds.pixel_mask = (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_9__.ones)([dims[0], dims[2], dims[3]]);
+    }
+    
     return await sessionRun(session, encoderFeeds);
 }
 
@@ -9843,6 +9899,7 @@ class PreTrainedModel extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_3__.Cal
                 if (this.config.model_type === 'musicgen') {
                     // Custom logic (TODO: move to Musicgen class)
                     decoder_input_ids = Array.from({
+                        // @ts-expect-error TS2339
                         length: batch_size * this.config.decoder.num_codebooks
                     }, () => [decoder_start_token_id]);
 
@@ -10172,11 +10229,13 @@ class PreTrainedModel extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_3__.Cal
     async encode_image({ pixel_values }) {
         // image_inputs === { pixel_values }
         const features = (await sessionRun(this.sessions['vision_encoder'], { pixel_values })).image_features;
+        // @ts-expect-error TS2339
         if (!this.config.num_image_tokens) {
             console.warn(
                 'The number of image tokens was not set in the model configuration. ' +
                 `Setting it to the number of features detected by the vision encoder (${features.dims[1]}).`
             )
+            // @ts-expect-error TS2339
             this.config.num_image_tokens = features.dims[1];
         }
         return features;
@@ -11604,6 +11663,7 @@ class WhisperForConditionalGeneration extends WhisperPreTrainedModel {
 
         if (generation_config.return_token_timestamps) {
             outputs["token_timestamps"] = this._extract_token_timestamps(
+                // @ts-expect-error TS2345
                 outputs,
                 generation_config.alignment_heads,
                 generation_config.num_frames,
@@ -11639,6 +11699,7 @@ class WhisperForConditionalGeneration extends WhisperPreTrainedModel {
             );
         }
 
+        // @ts-expect-error TS2339
         let median_filter_width = this.config.median_filter_width;
         if (median_filter_width === undefined) {
             console.warn("Model config has no `median_filter_width`, using default value of 7.")
@@ -11649,6 +11710,7 @@ class WhisperForConditionalGeneration extends WhisperPreTrainedModel {
         const batch = generate_outputs.cross_attentions;
         // Create a list with `decoder_layers` elements, each a tensor of shape
         // (batch size, attention_heads, output length, input length).
+        // @ts-expect-error TS2339
         const cross_attentions = Array.from({ length: this.config.decoder_layers },
             // Concatenate the cross attentions for each layer across sequence length dimension.
             (_, i) => (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_9__.cat)(batch.map(x => x[i]), 2)
@@ -11792,6 +11854,7 @@ class LlavaForConditionalGeneration extends LlavaPreTrainedModel {
         attention_mask,
     }) {
 
+        // @ts-expect-error TS2339
         const image_token_index = this.config.image_token_index;
 
         const idsList = input_ids.tolist();
@@ -12777,6 +12840,7 @@ class Qwen2VLForConditionalGeneration extends Qwen2VLPreTrainedModel {
                 const image_nums = vision_tokens.filter(x => x == image_token_id).length;
                 const video_nums = vision_tokens.filter(x => x == video_token_id).length;
 
+                /** @type {number[][]} */
                 let llm_pos_ids_list = [];
                 let st = 0;
                 let remain_images = image_nums;
@@ -12846,6 +12910,7 @@ class Qwen2VLForConditionalGeneration extends Qwen2VLPreTrainedModel {
                 // NOTE: Each item in llm_pos_ids_list is an array of shape (3, text_len),
                 // meaning to perform concatenation along dim=1, we can do the following:
                 const num_items = llm_pos_ids_list.reduce((acc, x) => acc + x.length, 0);
+                /** @type {number[]} */
                 const llm_positions = new Array(num_items);
                 let index = 0;
                 for (let x = 0; x < 3; ++x) {
@@ -12886,9 +12951,10 @@ class Qwen2VLForConditionalGeneration extends Qwen2VLPreTrainedModel {
                     { length: 3 * data.length },
                     (_, i) => data[i % data.length]
                 );
+                /** @type {bigint[]} */
                 const mrope_position_deltas = Array.from(
                     { length: dims[0] },
-                    (_, i) => (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_11__.max)(data.subarray(dims[1] * i, dims[1] * (i + 1)))[0] + 1 + dims[1]
+                    (_, i) => (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_11__.max)(data.subarray(dims[1] * i, dims[1] * (i + 1)))[0] + 1n + BigInt(dims[1])
                 );
 
                 return [
@@ -13459,7 +13525,7 @@ class DPTModel extends DPTPreTrainedModel { }
  * 
  * **Example:** Depth estimation w/ `Xenova/dpt-hybrid-midas`.
  * ```javascript
- * import { DPTForDepthEstimation, AutoProcessor, RawImage, interpolate, max } from '@huggingface/transformers';
+ * import { DPTForDepthEstimation, AutoProcessor, RawImage, interpolate_4d } from '@huggingface/transformers';
  * 
  * // Load model and processor
  * const model_id = 'Xenova/dpt-hybrid-midas';
@@ -13468,7 +13534,7 @@ class DPTModel extends DPTPreTrainedModel { }
  * 
  * // Load image from URL
  * const url = 'http://images.cocodataset.org/val2017/000000039769.jpg';
- * const image = await RawImage.fromURL(url);
+ * const image = await RawImage.read(url);
  * 
  * // Prepare image for the model
  * const inputs = await processor(image);
@@ -13477,10 +13543,15 @@ class DPTModel extends DPTPreTrainedModel { }
  * const { predicted_depth } = await model(inputs);
  * 
  * // Interpolate to original size
- * const prediction = interpolate(predicted_depth, image.size.reverse(), 'bilinear', false);
+ * const prediction = (await interpolate_4d(predicted_depth.unsqueeze(1), {
+     * size: image.size.reverse(),
+     * mode: 'bilinear',
+ * })).squeeze(1);
  * 
  * // Visualize the prediction
- * const formatted = prediction.mul_(255 / max(prediction.data)[0]).to('uint8');
+ * const min = prediction.min().item();
+ * const max = prediction.max().item();
+ * const formatted = prediction.sub_(min).div_(max - min).mul_(255).to('uint8');
  * const depth = RawImage.fromTensor(formatted);
  * // RawImage {
  * //   data: Uint8Array(307200) [ 85, 85, 84, ... ],
@@ -13530,11 +13601,7 @@ class GLPNPreTrainedModel extends PreTrainedModel { }
 class GLPNModel extends GLPNPreTrainedModel { }
 
 /**
- * GLPN Model transformer with a lightweight depth estimation head on top e.g. for KITTI, NYUv2.
- * 
- * **Example:** Depth estimation w/ `Xenova/glpn-kitti`.
- * ```javascript
- * import { GLPNForDepthEstimation, AutoProcessor, RawImage, interpolate, max } from '@huggingface/transformers';
+ * import { GLPNForDepthEstimation, AutoProcessor, RawImage, interpolate_4d } from '@huggingface/transformers';
  * 
  * // Load model and processor
  * const model_id = 'Xenova/glpn-kitti';
@@ -13543,7 +13610,7 @@ class GLPNModel extends GLPNPreTrainedModel { }
  * 
  * // Load image from URL
  * const url = 'http://images.cocodataset.org/val2017/000000039769.jpg';
- * const image = await RawImage.fromURL(url);
+ * const image = await RawImage.read(url);
  * 
  * // Prepare image for the model
  * const inputs = await processor(image);
@@ -13552,13 +13619,18 @@ class GLPNModel extends GLPNPreTrainedModel { }
  * const { predicted_depth } = await model(inputs);
  * 
  * // Interpolate to original size
- * const prediction = interpolate(predicted_depth, image.size.reverse(), 'bilinear', false);
+ * const prediction = (await interpolate_4d(predicted_depth.unsqueeze(1), {
+     * size: image.size.reverse(),
+     * mode: 'bilinear',
+ * })).squeeze(1);
  * 
  * // Visualize the prediction
- * const formatted = prediction.mul_(255 / max(prediction.data)[0]).to('uint8');
+ * const min = prediction.min().item();
+ * const max = prediction.max().item();
+ * const formatted = prediction.sub_(min).div_(max - min).mul_(255).to('uint8');
  * const depth = RawImage.fromTensor(formatted);
  * // RawImage {
- * //   data: Uint8Array(307200) [ 207, 169, 154, ... ],
+ * //   data: Uint8Array(307200) [ 85, 85, 84, ... ],
  * //   width: 640,
  * //   height: 480,
  * //   channels: 1
@@ -13733,6 +13805,8 @@ class Dinov2WithRegistersForImageClassification extends Dinov2WithRegistersPreTr
     }
 }
 //////////////////////////////////////////////////
+class GroundingDinoPreTrainedModel extends PreTrainedModel { }
+class GroundingDinoForObjectDetection extends GroundingDinoPreTrainedModel { }
 
 //////////////////////////////////////////////////
 class YolosPreTrainedModel extends PreTrainedModel { }
@@ -14431,6 +14505,9 @@ class WavLMForAudioFrameClassification extends WavLMPreTrainedModel {
     }
 }
 
+class StyleTextToSpeech2PreTrainedModel extends PreTrainedModel { }
+class StyleTextToSpeech2Model extends StyleTextToSpeech2PreTrainedModel { }
+
 //////////////////////////////////////////////////
 // SpeechT5 models
 /**
@@ -14525,10 +14602,12 @@ class SpeechT5ForTextToSpeech extends SpeechT5PreTrainedModel {
 
         const { encoder_outputs, encoder_attention_mask } = await encoderForward(this, model_inputs);
 
+        // @ts-expect-error TS2339
         const r = encoder_outputs.dims[1] / this.config.reduction_factor;
         const maxlen = Math.floor(r * maxlenratio);
         const minlen = Math.floor(r * minlenratio);
 
+        // @ts-expect-error TS2339
         const num_mel_bins = this.config.num_mel_bins;
 
         let spectrogramParts = [];
@@ -14893,11 +14972,13 @@ class MusicgenForConditionalGeneration extends PreTrainedModel { // NOTE: not Mu
      */
     _apply_and_filter_by_delay_pattern_mask(outputs) {
         const [bs_x_codebooks, seqLength] = outputs.dims;
+        // @ts-expect-error TS2339
         const num_codebooks = this.config.decoder.num_codebooks;
         const upperBound = (seqLength - num_codebooks);
 
         let newDataSize = 0;
         for (let i = 0; i < outputs.size; ++i) {
+            // @ts-expect-error TS2339
             if (outputs.data[i] === this.config.decoder.pad_token_id) {
                 continue;
             }
@@ -14927,7 +15008,9 @@ class MusicgenForConditionalGeneration extends PreTrainedModel { // NOTE: not Mu
         let clonedInputIds = structuredClone(input_ids);
         for (let i = 0; i < clonedInputIds.length; ++i) {
             for (let j = 0; j < clonedInputIds[i].length; ++j) {
+                // @ts-expect-error TS2339
                 if ((i % this.config.decoder.num_codebooks) >= j) {
+                    // @ts-expect-error TS2339
                     clonedInputIds[i][j] = BigInt(this.config.decoder.pad_token_id);
                 }
             }
@@ -15084,6 +15167,9 @@ class MultiModalityCausalLM extends MultiModalityPreTrainedModel {
         'past_key_values',
     ];
 
+    /**
+     * @param {ConstructorParameters<typeof MultiModalityPreTrainedModel>} args
+     */
     constructor(...args) {
         super(...args);
 
@@ -15385,6 +15471,8 @@ const MODEL_MAPPING_NAMES_ENCODER_ONLY = new Map([
 
     ['maskformer', ['MaskFormerModel', MaskFormerModel]],
     ['mgp-str', ['MgpstrForSceneTextRecognition', MgpstrForSceneTextRecognition]],
+
+    ['style_text_to_speech_2', ['StyleTextToSpeech2Model', StyleTextToSpeech2Model]],
 ]);
 
 const MODEL_MAPPING_NAMES_ENCODER_DECODER = new Map([
@@ -15629,6 +15717,7 @@ const MODEL_FOR_OBJECT_DETECTION_MAPPING_NAMES = new Map([
 const MODEL_FOR_ZERO_SHOT_OBJECT_DETECTION_MAPPING_NAMES = new Map([
     ['owlvit', ['OwlViTForObjectDetection', OwlViTForObjectDetection]],
     ['owlv2', ['Owlv2ForObjectDetection', Owlv2ForObjectDetection]],
+    ['grounding-dino', ['GroundingDinoForObjectDetection', GroundingDinoForObjectDetection]],
 ]);
 
 const MODEL_FOR_IMAGE_SEGMENTATION_MAPPING_NAMES = new Map([
@@ -16052,10 +16141,17 @@ class SequenceClassifierOutput extends ModelOutput {
     /**
      * @param {Object} output The output of the model.
      * @param {Tensor} output.logits classification (or regression if config.num_labels==1) scores (before SoftMax).
+     * @param {Record<string, Tensor>} [output.attentions] Object of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length, sequence_length)`.
+     * Attentions weights after the attention softmax, used to compute the weighted average in the self-attention heads.
      */
-    constructor({ logits }) {
+    constructor({ logits, ...attentions }) {
         super();
         this.logits = logits;
+        const attentions_list = Object.values(attentions);
+        if (attentions_list.length > 0) {
+            // Only set attentions if they are not empty
+            this.attentions = attentions_list;
+        }
     }
 }
 
@@ -16313,22 +16409,6 @@ __webpack_require__.r(__webpack_exports__);
 
 class AutoFeatureExtractor {
 
-    /**
-     * Instantiate one of the feature extractor classes of the library from a pretrained model.
-     * 
-     * The processor class to instantiate is selected based on the `feature_extractor_type` property of
-     * the config object (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
-     * 
-     * @param {string} pretrained_model_name_or_path The name or path of the pretrained model. Can be either:
-     * - A string, the *model id* of a pretrained processor hosted inside a model repo on huggingface.co.
-     *   Valid model ids can be located at the root-level, like `bert-base-uncased`, or namespaced under a
-     *   user or organization name, like `dbmdz/bert-base-german-cased`.
-     * - A path to a *directory* containing processor files, e.g., `./my_model_directory/`.
-     * @param {import('../../utils/hub.js').PretrainedOptions} options Additional options for loading the processor.
-     * 
-     * @returns {Promise<AllFeatureExtractors.ImageProcessor>} A new instance of the Processor class.
-     */
-
     /** @type {typeof FeatureExtractor.from_pretrained} */
     static async from_pretrained(pretrained_model_name_or_path, options={}) {
 
@@ -16457,22 +16537,6 @@ __webpack_require__.r(__webpack_exports__);
  */
 class AutoProcessor {
 
-    /**
-     * Instantiate one of the processor classes of the library from a pretrained model.
-     * 
-     * The processor class to instantiate is selected based on the `image_processor_type` (or `feature_extractor_type`; legacy)
-     * property of the config object (either passed as an argument or loaded from `pretrained_model_name_or_path` if possible)
-     * 
-     * @param {string} pretrained_model_name_or_path The name or path of the pretrained model. Can be either:
-     * - A string, the *model id* of a pretrained processor hosted inside a model repo on huggingface.co.
-     *   Valid model ids can be located at the root-level, like `bert-base-uncased`, or namespaced under a
-     *   user or organization name, like `dbmdz/bert-base-german-cased`.
-     * - A path to a *directory* containing processor files, e.g., `./my_model_directory/`.
-     * @param {import('../../utils/hub.js').PretrainedOptions} options Additional options for loading the processor.
-     * 
-     * @returns {Promise<Processor>} A new instance of the Processor class.
-     */
-
     /** @type {typeof Processor.from_pretrained} */
     static async from_pretrained(pretrained_model_name_or_path, options={}) {
 
@@ -16796,6 +16860,7 @@ class ConvNextImageProcessor extends _base_image_processors_utils_js__WEBPACK_IM
         /**
          * Percentage of the image to crop. Only has an effect if this.size < 384.
          */
+        // @ts-expect-error TS2339
         this.crop_pct = this.config.crop_pct ?? (224 / 256);
     }
 
@@ -17003,6 +17068,7 @@ __webpack_require__.r(__webpack_exports__);
 class EfficientNetImageProcessor extends _base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_0__.ImageProcessor {
     constructor(config) {
         super(config);
+        // @ts-expect-error TS2339
         this.include_top = this.config.include_top ?? true;
         if (this.include_top) {
             this.image_std = this.image_std.map(x => x * x);
@@ -17086,8 +17152,11 @@ class Florence2Processor extends _base_processing_utils_js__WEBPACK_IMPORTED_MOD
         super(config, components);
 
         const {
+            // @ts-expect-error TS2339
             tasks_answer_post_processing_type,
+            // @ts-expect-error TS2339
             task_prompts_without_inputs,
+            // @ts-expect-error TS2339
             task_prompts_with_input,
         } = this.image_processor.config;
 
@@ -17223,6 +17292,170 @@ __webpack_require__.r(__webpack_exports__);
 class GLPNFeatureExtractor extends _base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_0__.ImageProcessor { }
 
 
+/***/ }),
+
+/***/ "./src/models/grounding_dino/image_processing_grounding_dino.js":
+/*!**********************************************************************!*\
+  !*** ./src/models/grounding_dino/image_processing_grounding_dino.js ***!
+  \**********************************************************************/
+/***/ ((__unused_webpack___webpack_module__, __webpack_exports__, __webpack_require__) => {
+
+"use strict";
+__webpack_require__.r(__webpack_exports__);
+/* harmony export */ __webpack_require__.d(__webpack_exports__, {
+/* harmony export */   GroundingDinoImageProcessor: () => (/* binding */ GroundingDinoImageProcessor)
+/* harmony export */ });
+/* harmony import */ var _base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ../../base/image_processors_utils.js */ "./src/base/image_processors_utils.js");
+/* harmony import */ var _utils_tensor_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ../../utils/tensor.js */ "./src/utils/tensor.js");
+
+
+
+
+
+/**
+ * @typedef {object} GroundingDinoFeatureExtractorResultProps
+ * @property {import('../../utils/tensor.js').Tensor} pixel_mask
+ * @typedef {import('../../base/image_processors_utils.js').ImageProcessorResult & GroundingDinoFeatureExtractorResultProps} GroundingDinoFeatureExtractorResult
+ */
+
+class GroundingDinoImageProcessor extends _base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_0__.ImageProcessor {
+    /**
+     * Calls the feature extraction process on an array of images, preprocesses
+     * each image, and concatenates the resulting features into a single Tensor.
+     * @param {import('../../utils/image.js').RawImage[]} images The image(s) to extract features from.
+     * @returns {Promise<GroundingDinoFeatureExtractorResult>} An object containing the concatenated pixel values of the preprocessed images.
+     */
+    async _call(images) {
+        const result = await super._call(images);
+
+        const dims = result.pixel_values.dims;
+        const pixel_mask = (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_1__.ones)([dims[0], dims[2], dims[3]]);
+
+        return { ...result, pixel_mask };
+    }
+}
+
+
+/***/ }),
+
+/***/ "./src/models/grounding_dino/processing_grounding_dino.js":
+/*!****************************************************************!*\
+  !*** ./src/models/grounding_dino/processing_grounding_dino.js ***!
+  \****************************************************************/
+/***/ ((__unused_webpack___webpack_module__, __webpack_exports__, __webpack_require__) => {
+
+"use strict";
+__webpack_require__.r(__webpack_exports__);
+/* harmony export */ __webpack_require__.d(__webpack_exports__, {
+/* harmony export */   GroundingDinoProcessor: () => (/* binding */ GroundingDinoProcessor)
+/* harmony export */ });
+/* harmony import */ var _base_processing_utils_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ../../base/processing_utils.js */ "./src/base/processing_utils.js");
+/* harmony import */ var _auto_image_processing_auto_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ../auto/image_processing_auto.js */ "./src/models/auto/image_processing_auto.js");
+/* harmony import */ var _tokenizers_js__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! ../../tokenizers.js */ "./src/tokenizers.js");
+/* harmony import */ var _base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! ../../base/image_processors_utils.js */ "./src/base/image_processors_utils.js");
+
+
+
+
+
+/**
+ * Get token ids of phrases from posmaps and input_ids.
+ * @param {import('../../utils/tensor.js').Tensor} posmaps A boolean tensor of unbatched text-thresholded logits related to the detected bounding boxes of shape `(hidden_size, )`.
+ * @param {import('../../utils/tensor.js').Tensor} input_ids A tensor of token ids of shape `(sequence_length, )`.
+ */
+function get_phrases_from_posmap(posmaps, input_ids) {
+
+    const left_idx = 0;
+    const right_idx = posmaps.dims.at(-1) - 1;
+
+    const posmaps_list = posmaps.tolist();
+    posmaps_list.fill(false, 0, left_idx + 1);
+    posmaps_list.fill(false, right_idx);
+
+    const input_ids_list = input_ids.tolist();
+    return posmaps_list
+        .map((val, idx) => val ? idx : null)
+        .filter(idx => idx !== null)
+        .map(i => input_ids_list[i]);
+}
+
+class GroundingDinoProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODULE_0__.Processor {
+    static tokenizer_class = _tokenizers_js__WEBPACK_IMPORTED_MODULE_2__.AutoTokenizer
+    static image_processor_class = _auto_image_processing_auto_js__WEBPACK_IMPORTED_MODULE_1__.AutoImageProcessor
+
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     */
+    /**
+     * 
+     * @param {RawImage|RawImage[]|RawImage[][]} images  
+     * @param {string|string[]} text 
+     * @returns {Promise<any>}
+     */
+    async _call(images, text, options = {}) {
+
+        const image_inputs = images ? await this.image_processor(images, options) : {};
+        const text_inputs = text ? this.tokenizer(text, options) : {};
+
+        return {
+            ...text_inputs,
+            ...image_inputs,
+        }
+    }
+    post_process_grounded_object_detection(outputs, input_ids, {
+        box_threshold = 0.25,
+        text_threshold = 0.25,
+        target_sizes = null
+    } = {}) {
+        const { logits, pred_boxes } = outputs;
+        const batch_size = logits.dims[0];
+
+        if (target_sizes !== null && target_sizes.length !== batch_size) {
+            throw Error("Make sure that you pass in as many target sizes as the batch dimension of the logits")
+        }
+        const num_queries = logits.dims.at(1);
+
+        const probs = logits.sigmoid(); // (batch_size, num_queries, 256)
+        const scores = probs.max(-1).tolist(); // (batch_size, num_queries)
+
+        // Convert to [x0, y0, x1, y1] format
+        const boxes = pred_boxes.tolist() // (batch_size, num_queries, 4)
+            .map(batch => batch.map(box => (0,_base_image_processors_utils_js__WEBPACK_IMPORTED_MODULE_3__.center_to_corners_format)(box)));
+
+        const results = [];
+        for (let i = 0; i < batch_size; ++i) {
+            const target_size = target_sizes !== null ? target_sizes[i] : null;
+
+            // Convert from relative [0, 1] to absolute [0, height] coordinates
+            if (target_size !== null) {
+                boxes[i] = boxes[i].map(box => box.map((x, j) => x * target_size[(j + 1) % 2]));
+            }
+
+            const batch_scores = scores[i];
+            const final_scores = [];
+            const final_phrases = [];
+            const final_boxes = [];
+            for (let j = 0; j < num_queries; ++j) {
+                const score = batch_scores[j];
+                if (score <= box_threshold) {
+                    continue;
+                }
+                const box = boxes[i][j];
+                const prob = probs[i][j];
+
+                final_scores.push(score);
+                final_boxes.push(box);
+
+                const phrases = get_phrases_from_posmap(prob.gt(text_threshold), input_ids[i]);
+                final_phrases.push(phrases);
+            }
+            results.push({ scores: final_scores, boxes: final_boxes, labels: this.batch_decode(final_phrases) });
+        }
+        return results;
+    }
+}
+
+
 /***/ }),
 
 /***/ "./src/models/idefics3/image_processing_idefics3.js":
@@ -17384,6 +17617,8 @@ class Idefics3ImageProcessor extends _base_image_processors_utils_js__WEBPACK_IM
 
                     const start_offset = i * pixel_attention_mask_stride + num_patches * h * w;
                     const end_offset = (i + 1) * pixel_attention_mask_stride;
+
+                    // @ts-expect-error
                     pixel_attention_mask_data.fill(false, start_offset, end_offset);
                 }
             }
@@ -17652,42 +17887,43 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   DonutImageProcessor: () => (/* reexport safe */ _donut_image_processing_donut_js__WEBPACK_IMPORTED_MODULE_7__.DonutImageProcessor),
 /* harmony export */   EfficientNetImageProcessor: () => (/* reexport safe */ _efficientnet_image_processing_efficientnet_js__WEBPACK_IMPORTED_MODULE_9__.EfficientNetImageProcessor),
 /* harmony export */   GLPNFeatureExtractor: () => (/* reexport safe */ _glpn_image_processing_glpn_js__WEBPACK_IMPORTED_MODULE_10__.GLPNFeatureExtractor),
-/* harmony export */   Idefics3ImageProcessor: () => (/* reexport safe */ _idefics3_image_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_11__.Idefics3ImageProcessor),
-/* harmony export */   JinaCLIPImageProcessor: () => (/* reexport safe */ _jina_clip_image_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_13__.JinaCLIPImageProcessor),
-/* harmony export */   LlavaOnevisionImageProcessor: () => (/* reexport safe */ _llava_onevision_image_processing_llava_onevision_js__WEBPACK_IMPORTED_MODULE_14__.LlavaOnevisionImageProcessor),
-/* harmony export */   Mask2FormerImageProcessor: () => (/* reexport safe */ _mask2former_image_processing_mask2former_js__WEBPACK_IMPORTED_MODULE_15__.Mask2FormerImageProcessor),
-/* harmony export */   MaskFormerFeatureExtractor: () => (/* reexport safe */ _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_16__.MaskFormerFeatureExtractor),
-/* harmony export */   MaskFormerImageProcessor: () => (/* reexport safe */ _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_16__.MaskFormerImageProcessor),
-/* harmony export */   MobileNetV1FeatureExtractor: () => (/* reexport safe */ _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_17__.MobileNetV1FeatureExtractor),
-/* harmony export */   MobileNetV1ImageProcessor: () => (/* reexport safe */ _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_17__.MobileNetV1ImageProcessor),
-/* harmony export */   MobileNetV2FeatureExtractor: () => (/* reexport safe */ _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_18__.MobileNetV2FeatureExtractor),
-/* harmony export */   MobileNetV2ImageProcessor: () => (/* reexport safe */ _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_18__.MobileNetV2ImageProcessor),
-/* harmony export */   MobileNetV3FeatureExtractor: () => (/* reexport safe */ _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_19__.MobileNetV3FeatureExtractor),
-/* harmony export */   MobileNetV3ImageProcessor: () => (/* reexport safe */ _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_19__.MobileNetV3ImageProcessor),
-/* harmony export */   MobileNetV4FeatureExtractor: () => (/* reexport safe */ _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_20__.MobileNetV4FeatureExtractor),
-/* harmony export */   MobileNetV4ImageProcessor: () => (/* reexport safe */ _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_20__.MobileNetV4ImageProcessor),
-/* harmony export */   MobileViTFeatureExtractor: () => (/* reexport safe */ _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_21__.MobileViTFeatureExtractor),
-/* harmony export */   MobileViTImageProcessor: () => (/* reexport safe */ _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_21__.MobileViTImageProcessor),
-/* harmony export */   NougatImageProcessor: () => (/* reexport safe */ _nougat_image_processing_nougat_js__WEBPACK_IMPORTED_MODULE_22__.NougatImageProcessor),
-/* harmony export */   OwlViTFeatureExtractor: () => (/* reexport safe */ _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_24__.OwlViTFeatureExtractor),
-/* harmony export */   OwlViTImageProcessor: () => (/* reexport safe */ _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_24__.OwlViTImageProcessor),
-/* harmony export */   Owlv2ImageProcessor: () => (/* reexport safe */ _owlv2_image_processing_owlv2_js__WEBPACK_IMPORTED_MODULE_23__.Owlv2ImageProcessor),
-/* harmony export */   Phi3VImageProcessor: () => (/* reexport safe */ _phi3_v_image_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_25__.Phi3VImageProcessor),
-/* harmony export */   PvtImageProcessor: () => (/* reexport safe */ _pvt_image_processing_pvt_js__WEBPACK_IMPORTED_MODULE_26__.PvtImageProcessor),
-/* harmony export */   Qwen2VLImageProcessor: () => (/* reexport safe */ _qwen2_vl_image_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_27__.Qwen2VLImageProcessor),
-/* harmony export */   RTDetrImageProcessor: () => (/* reexport safe */ _rt_detr_image_processing_rt_detr_js__WEBPACK_IMPORTED_MODULE_28__.RTDetrImageProcessor),
-/* harmony export */   SamImageProcessor: () => (/* reexport safe */ _sam_image_processing_sam_js__WEBPACK_IMPORTED_MODULE_29__.SamImageProcessor),
-/* harmony export */   SegformerFeatureExtractor: () => (/* reexport safe */ _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_30__.SegformerFeatureExtractor),
-/* harmony export */   SegformerImageProcessor: () => (/* reexport safe */ _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_30__.SegformerImageProcessor),
-/* harmony export */   SiglipImageProcessor: () => (/* reexport safe */ _siglip_image_processing_siglip_js__WEBPACK_IMPORTED_MODULE_31__.SiglipImageProcessor),
-/* harmony export */   Swin2SRImageProcessor: () => (/* reexport safe */ _swin2sr_image_processing_swin2sr_js__WEBPACK_IMPORTED_MODULE_32__.Swin2SRImageProcessor),
-/* harmony export */   VLMImageProcessor: () => (/* reexport safe */ _janus_image_processing_janus_js__WEBPACK_IMPORTED_MODULE_12__.VLMImageProcessor),
-/* harmony export */   ViTFeatureExtractor: () => (/* reexport safe */ _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_33__.ViTFeatureExtractor),
-/* harmony export */   ViTImageProcessor: () => (/* reexport safe */ _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_33__.ViTImageProcessor),
-/* harmony export */   VitMatteImageProcessor: () => (/* reexport safe */ _vitmatte_image_processing_vitmatte_js__WEBPACK_IMPORTED_MODULE_34__.VitMatteImageProcessor),
-/* harmony export */   VitPoseImageProcessor: () => (/* reexport safe */ _vitpose_image_processing_vitpose_js__WEBPACK_IMPORTED_MODULE_35__.VitPoseImageProcessor),
-/* harmony export */   YolosFeatureExtractor: () => (/* reexport safe */ _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_36__.YolosFeatureExtractor),
-/* harmony export */   YolosImageProcessor: () => (/* reexport safe */ _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_36__.YolosImageProcessor)
+/* harmony export */   GroundingDinoImageProcessor: () => (/* reexport safe */ _grounding_dino_image_processing_grounding_dino_js__WEBPACK_IMPORTED_MODULE_11__.GroundingDinoImageProcessor),
+/* harmony export */   Idefics3ImageProcessor: () => (/* reexport safe */ _idefics3_image_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_12__.Idefics3ImageProcessor),
+/* harmony export */   JinaCLIPImageProcessor: () => (/* reexport safe */ _jina_clip_image_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_14__.JinaCLIPImageProcessor),
+/* harmony export */   LlavaOnevisionImageProcessor: () => (/* reexport safe */ _llava_onevision_image_processing_llava_onevision_js__WEBPACK_IMPORTED_MODULE_15__.LlavaOnevisionImageProcessor),
+/* harmony export */   Mask2FormerImageProcessor: () => (/* reexport safe */ _mask2former_image_processing_mask2former_js__WEBPACK_IMPORTED_MODULE_16__.Mask2FormerImageProcessor),
+/* harmony export */   MaskFormerFeatureExtractor: () => (/* reexport safe */ _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_17__.MaskFormerFeatureExtractor),
+/* harmony export */   MaskFormerImageProcessor: () => (/* reexport safe */ _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_17__.MaskFormerImageProcessor),
+/* harmony export */   MobileNetV1FeatureExtractor: () => (/* reexport safe */ _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_18__.MobileNetV1FeatureExtractor),
+/* harmony export */   MobileNetV1ImageProcessor: () => (/* reexport safe */ _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_18__.MobileNetV1ImageProcessor),
+/* harmony export */   MobileNetV2FeatureExtractor: () => (/* reexport safe */ _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_19__.MobileNetV2FeatureExtractor),
+/* harmony export */   MobileNetV2ImageProcessor: () => (/* reexport safe */ _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_19__.MobileNetV2ImageProcessor),
+/* harmony export */   MobileNetV3FeatureExtractor: () => (/* reexport safe */ _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_20__.MobileNetV3FeatureExtractor),
+/* harmony export */   MobileNetV3ImageProcessor: () => (/* reexport safe */ _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_20__.MobileNetV3ImageProcessor),
+/* harmony export */   MobileNetV4FeatureExtractor: () => (/* reexport safe */ _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_21__.MobileNetV4FeatureExtractor),
+/* harmony export */   MobileNetV4ImageProcessor: () => (/* reexport safe */ _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_21__.MobileNetV4ImageProcessor),
+/* harmony export */   MobileViTFeatureExtractor: () => (/* reexport safe */ _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_22__.MobileViTFeatureExtractor),
+/* harmony export */   MobileViTImageProcessor: () => (/* reexport safe */ _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_22__.MobileViTImageProcessor),
+/* harmony export */   NougatImageProcessor: () => (/* reexport safe */ _nougat_image_processing_nougat_js__WEBPACK_IMPORTED_MODULE_23__.NougatImageProcessor),
+/* harmony export */   OwlViTFeatureExtractor: () => (/* reexport safe */ _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_25__.OwlViTFeatureExtractor),
+/* harmony export */   OwlViTImageProcessor: () => (/* reexport safe */ _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_25__.OwlViTImageProcessor),
+/* harmony export */   Owlv2ImageProcessor: () => (/* reexport safe */ _owlv2_image_processing_owlv2_js__WEBPACK_IMPORTED_MODULE_24__.Owlv2ImageProcessor),
+/* harmony export */   Phi3VImageProcessor: () => (/* reexport safe */ _phi3_v_image_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_26__.Phi3VImageProcessor),
+/* harmony export */   PvtImageProcessor: () => (/* reexport safe */ _pvt_image_processing_pvt_js__WEBPACK_IMPORTED_MODULE_27__.PvtImageProcessor),
+/* harmony export */   Qwen2VLImageProcessor: () => (/* reexport safe */ _qwen2_vl_image_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_28__.Qwen2VLImageProcessor),
+/* harmony export */   RTDetrImageProcessor: () => (/* reexport safe */ _rt_detr_image_processing_rt_detr_js__WEBPACK_IMPORTED_MODULE_29__.RTDetrImageProcessor),
+/* harmony export */   SamImageProcessor: () => (/* reexport safe */ _sam_image_processing_sam_js__WEBPACK_IMPORTED_MODULE_30__.SamImageProcessor),
+/* harmony export */   SegformerFeatureExtractor: () => (/* reexport safe */ _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_31__.SegformerFeatureExtractor),
+/* harmony export */   SegformerImageProcessor: () => (/* reexport safe */ _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_31__.SegformerImageProcessor),
+/* harmony export */   SiglipImageProcessor: () => (/* reexport safe */ _siglip_image_processing_siglip_js__WEBPACK_IMPORTED_MODULE_32__.SiglipImageProcessor),
+/* harmony export */   Swin2SRImageProcessor: () => (/* reexport safe */ _swin2sr_image_processing_swin2sr_js__WEBPACK_IMPORTED_MODULE_33__.Swin2SRImageProcessor),
+/* harmony export */   VLMImageProcessor: () => (/* reexport safe */ _janus_image_processing_janus_js__WEBPACK_IMPORTED_MODULE_13__.VLMImageProcessor),
+/* harmony export */   ViTFeatureExtractor: () => (/* reexport safe */ _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_34__.ViTFeatureExtractor),
+/* harmony export */   ViTImageProcessor: () => (/* reexport safe */ _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_34__.ViTImageProcessor),
+/* harmony export */   VitMatteImageProcessor: () => (/* reexport safe */ _vitmatte_image_processing_vitmatte_js__WEBPACK_IMPORTED_MODULE_35__.VitMatteImageProcessor),
+/* harmony export */   VitPoseImageProcessor: () => (/* reexport safe */ _vitpose_image_processing_vitpose_js__WEBPACK_IMPORTED_MODULE_36__.VitPoseImageProcessor),
+/* harmony export */   YolosFeatureExtractor: () => (/* reexport safe */ _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_37__.YolosFeatureExtractor),
+/* harmony export */   YolosImageProcessor: () => (/* reexport safe */ _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_37__.YolosImageProcessor)
 /* harmony export */ });
 /* harmony import */ var _beit_image_processing_beit_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ./beit/image_processing_beit.js */ "./src/models/beit/image_processing_beit.js");
 /* harmony import */ var _bit_image_processing_bit_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ./bit/image_processing_bit.js */ "./src/models/bit/image_processing_bit.js");
@@ -17700,32 +17936,34 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony import */ var _dpt_image_processing_dpt_js__WEBPACK_IMPORTED_MODULE_8__ = __webpack_require__(/*! ./dpt/image_processing_dpt.js */ "./src/models/dpt/image_processing_dpt.js");
 /* harmony import */ var _efficientnet_image_processing_efficientnet_js__WEBPACK_IMPORTED_MODULE_9__ = __webpack_require__(/*! ./efficientnet/image_processing_efficientnet.js */ "./src/models/efficientnet/image_processing_efficientnet.js");
 /* harmony import */ var _glpn_image_processing_glpn_js__WEBPACK_IMPORTED_MODULE_10__ = __webpack_require__(/*! ./glpn/image_processing_glpn.js */ "./src/models/glpn/image_processing_glpn.js");
-/* harmony import */ var _idefics3_image_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./idefics3/image_processing_idefics3.js */ "./src/models/idefics3/image_processing_idefics3.js");
-/* harmony import */ var _janus_image_processing_janus_js__WEBPACK_IMPORTED_MODULE_12__ = __webpack_require__(/*! ./janus/image_processing_janus.js */ "./src/models/janus/image_processing_janus.js");
-/* harmony import */ var _jina_clip_image_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_13__ = __webpack_require__(/*! ./jina_clip/image_processing_jina_clip.js */ "./src/models/jina_clip/image_processing_jina_clip.js");
-/* harmony import */ var _llava_onevision_image_processing_llava_onevision_js__WEBPACK_IMPORTED_MODULE_14__ = __webpack_require__(/*! ./llava_onevision/image_processing_llava_onevision.js */ "./src/models/llava_onevision/image_processing_llava_onevision.js");
-/* harmony import */ var _mask2former_image_processing_mask2former_js__WEBPACK_IMPORTED_MODULE_15__ = __webpack_require__(/*! ./mask2former/image_processing_mask2former.js */ "./src/models/mask2former/image_processing_mask2former.js");
-/* harmony import */ var _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_16__ = __webpack_require__(/*! ./maskformer/image_processing_maskformer.js */ "./src/models/maskformer/image_processing_maskformer.js");
-/* harmony import */ var _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_17__ = __webpack_require__(/*! ./mobilenet_v1/image_processing_mobilenet_v1.js */ "./src/models/mobilenet_v1/image_processing_mobilenet_v1.js");
-/* harmony import */ var _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_18__ = __webpack_require__(/*! ./mobilenet_v2/image_processing_mobilenet_v2.js */ "./src/models/mobilenet_v2/image_processing_mobilenet_v2.js");
-/* harmony import */ var _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_19__ = __webpack_require__(/*! ./mobilenet_v3/image_processing_mobilenet_v3.js */ "./src/models/mobilenet_v3/image_processing_mobilenet_v3.js");
-/* harmony import */ var _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_20__ = __webpack_require__(/*! ./mobilenet_v4/image_processing_mobilenet_v4.js */ "./src/models/mobilenet_v4/image_processing_mobilenet_v4.js");
-/* harmony import */ var _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_21__ = __webpack_require__(/*! ./mobilevit/image_processing_mobilevit.js */ "./src/models/mobilevit/image_processing_mobilevit.js");
-/* harmony import */ var _nougat_image_processing_nougat_js__WEBPACK_IMPORTED_MODULE_22__ = __webpack_require__(/*! ./nougat/image_processing_nougat.js */ "./src/models/nougat/image_processing_nougat.js");
-/* harmony import */ var _owlv2_image_processing_owlv2_js__WEBPACK_IMPORTED_MODULE_23__ = __webpack_require__(/*! ./owlv2/image_processing_owlv2.js */ "./src/models/owlv2/image_processing_owlv2.js");
-/* harmony import */ var _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_24__ = __webpack_require__(/*! ./owlvit/image_processing_owlvit.js */ "./src/models/owlvit/image_processing_owlvit.js");
-/* harmony import */ var _phi3_v_image_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_25__ = __webpack_require__(/*! ./phi3_v/image_processing_phi3_v.js */ "./src/models/phi3_v/image_processing_phi3_v.js");
-/* harmony import */ var _pvt_image_processing_pvt_js__WEBPACK_IMPORTED_MODULE_26__ = __webpack_require__(/*! ./pvt/image_processing_pvt.js */ "./src/models/pvt/image_processing_pvt.js");
-/* harmony import */ var _qwen2_vl_image_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_27__ = __webpack_require__(/*! ./qwen2_vl/image_processing_qwen2_vl.js */ "./src/models/qwen2_vl/image_processing_qwen2_vl.js");
-/* harmony import */ var _rt_detr_image_processing_rt_detr_js__WEBPACK_IMPORTED_MODULE_28__ = __webpack_require__(/*! ./rt_detr/image_processing_rt_detr.js */ "./src/models/rt_detr/image_processing_rt_detr.js");
-/* harmony import */ var _sam_image_processing_sam_js__WEBPACK_IMPORTED_MODULE_29__ = __webpack_require__(/*! ./sam/image_processing_sam.js */ "./src/models/sam/image_processing_sam.js");
-/* harmony import */ var _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_30__ = __webpack_require__(/*! ./segformer/image_processing_segformer.js */ "./src/models/segformer/image_processing_segformer.js");
-/* harmony import */ var _siglip_image_processing_siglip_js__WEBPACK_IMPORTED_MODULE_31__ = __webpack_require__(/*! ./siglip/image_processing_siglip.js */ "./src/models/siglip/image_processing_siglip.js");
-/* harmony import */ var _swin2sr_image_processing_swin2sr_js__WEBPACK_IMPORTED_MODULE_32__ = __webpack_require__(/*! ./swin2sr/image_processing_swin2sr.js */ "./src/models/swin2sr/image_processing_swin2sr.js");
-/* harmony import */ var _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_33__ = __webpack_require__(/*! ./vit/image_processing_vit.js */ "./src/models/vit/image_processing_vit.js");
-/* harmony import */ var _vitmatte_image_processing_vitmatte_js__WEBPACK_IMPORTED_MODULE_34__ = __webpack_require__(/*! ./vitmatte/image_processing_vitmatte.js */ "./src/models/vitmatte/image_processing_vitmatte.js");
-/* harmony import */ var _vitpose_image_processing_vitpose_js__WEBPACK_IMPORTED_MODULE_35__ = __webpack_require__(/*! ./vitpose/image_processing_vitpose.js */ "./src/models/vitpose/image_processing_vitpose.js");
-/* harmony import */ var _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_36__ = __webpack_require__(/*! ./yolos/image_processing_yolos.js */ "./src/models/yolos/image_processing_yolos.js");
+/* harmony import */ var _grounding_dino_image_processing_grounding_dino_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./grounding_dino/image_processing_grounding_dino.js */ "./src/models/grounding_dino/image_processing_grounding_dino.js");
+/* harmony import */ var _idefics3_image_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_12__ = __webpack_require__(/*! ./idefics3/image_processing_idefics3.js */ "./src/models/idefics3/image_processing_idefics3.js");
+/* harmony import */ var _janus_image_processing_janus_js__WEBPACK_IMPORTED_MODULE_13__ = __webpack_require__(/*! ./janus/image_processing_janus.js */ "./src/models/janus/image_processing_janus.js");
+/* harmony import */ var _jina_clip_image_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_14__ = __webpack_require__(/*! ./jina_clip/image_processing_jina_clip.js */ "./src/models/jina_clip/image_processing_jina_clip.js");
+/* harmony import */ var _llava_onevision_image_processing_llava_onevision_js__WEBPACK_IMPORTED_MODULE_15__ = __webpack_require__(/*! ./llava_onevision/image_processing_llava_onevision.js */ "./src/models/llava_onevision/image_processing_llava_onevision.js");
+/* harmony import */ var _mask2former_image_processing_mask2former_js__WEBPACK_IMPORTED_MODULE_16__ = __webpack_require__(/*! ./mask2former/image_processing_mask2former.js */ "./src/models/mask2former/image_processing_mask2former.js");
+/* harmony import */ var _maskformer_image_processing_maskformer_js__WEBPACK_IMPORTED_MODULE_17__ = __webpack_require__(/*! ./maskformer/image_processing_maskformer.js */ "./src/models/maskformer/image_processing_maskformer.js");
+/* harmony import */ var _mobilenet_v1_image_processing_mobilenet_v1_js__WEBPACK_IMPORTED_MODULE_18__ = __webpack_require__(/*! ./mobilenet_v1/image_processing_mobilenet_v1.js */ "./src/models/mobilenet_v1/image_processing_mobilenet_v1.js");
+/* harmony import */ var _mobilenet_v2_image_processing_mobilenet_v2_js__WEBPACK_IMPORTED_MODULE_19__ = __webpack_require__(/*! ./mobilenet_v2/image_processing_mobilenet_v2.js */ "./src/models/mobilenet_v2/image_processing_mobilenet_v2.js");
+/* harmony import */ var _mobilenet_v3_image_processing_mobilenet_v3_js__WEBPACK_IMPORTED_MODULE_20__ = __webpack_require__(/*! ./mobilenet_v3/image_processing_mobilenet_v3.js */ "./src/models/mobilenet_v3/image_processing_mobilenet_v3.js");
+/* harmony import */ var _mobilenet_v4_image_processing_mobilenet_v4_js__WEBPACK_IMPORTED_MODULE_21__ = __webpack_require__(/*! ./mobilenet_v4/image_processing_mobilenet_v4.js */ "./src/models/mobilenet_v4/image_processing_mobilenet_v4.js");
+/* harmony import */ var _mobilevit_image_processing_mobilevit_js__WEBPACK_IMPORTED_MODULE_22__ = __webpack_require__(/*! ./mobilevit/image_processing_mobilevit.js */ "./src/models/mobilevit/image_processing_mobilevit.js");
+/* harmony import */ var _nougat_image_processing_nougat_js__WEBPACK_IMPORTED_MODULE_23__ = __webpack_require__(/*! ./nougat/image_processing_nougat.js */ "./src/models/nougat/image_processing_nougat.js");
+/* harmony import */ var _owlv2_image_processing_owlv2_js__WEBPACK_IMPORTED_MODULE_24__ = __webpack_require__(/*! ./owlv2/image_processing_owlv2.js */ "./src/models/owlv2/image_processing_owlv2.js");
+/* harmony import */ var _owlvit_image_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_25__ = __webpack_require__(/*! ./owlvit/image_processing_owlvit.js */ "./src/models/owlvit/image_processing_owlvit.js");
+/* harmony import */ var _phi3_v_image_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_26__ = __webpack_require__(/*! ./phi3_v/image_processing_phi3_v.js */ "./src/models/phi3_v/image_processing_phi3_v.js");
+/* harmony import */ var _pvt_image_processing_pvt_js__WEBPACK_IMPORTED_MODULE_27__ = __webpack_require__(/*! ./pvt/image_processing_pvt.js */ "./src/models/pvt/image_processing_pvt.js");
+/* harmony import */ var _qwen2_vl_image_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_28__ = __webpack_require__(/*! ./qwen2_vl/image_processing_qwen2_vl.js */ "./src/models/qwen2_vl/image_processing_qwen2_vl.js");
+/* harmony import */ var _rt_detr_image_processing_rt_detr_js__WEBPACK_IMPORTED_MODULE_29__ = __webpack_require__(/*! ./rt_detr/image_processing_rt_detr.js */ "./src/models/rt_detr/image_processing_rt_detr.js");
+/* harmony import */ var _sam_image_processing_sam_js__WEBPACK_IMPORTED_MODULE_30__ = __webpack_require__(/*! ./sam/image_processing_sam.js */ "./src/models/sam/image_processing_sam.js");
+/* harmony import */ var _segformer_image_processing_segformer_js__WEBPACK_IMPORTED_MODULE_31__ = __webpack_require__(/*! ./segformer/image_processing_segformer.js */ "./src/models/segformer/image_processing_segformer.js");
+/* harmony import */ var _siglip_image_processing_siglip_js__WEBPACK_IMPORTED_MODULE_32__ = __webpack_require__(/*! ./siglip/image_processing_siglip.js */ "./src/models/siglip/image_processing_siglip.js");
+/* harmony import */ var _swin2sr_image_processing_swin2sr_js__WEBPACK_IMPORTED_MODULE_33__ = __webpack_require__(/*! ./swin2sr/image_processing_swin2sr.js */ "./src/models/swin2sr/image_processing_swin2sr.js");
+/* harmony import */ var _vit_image_processing_vit_js__WEBPACK_IMPORTED_MODULE_34__ = __webpack_require__(/*! ./vit/image_processing_vit.js */ "./src/models/vit/image_processing_vit.js");
+/* harmony import */ var _vitmatte_image_processing_vitmatte_js__WEBPACK_IMPORTED_MODULE_35__ = __webpack_require__(/*! ./vitmatte/image_processing_vitmatte.js */ "./src/models/vitmatte/image_processing_vitmatte.js");
+/* harmony import */ var _vitpose_image_processing_vitpose_js__WEBPACK_IMPORTED_MODULE_36__ = __webpack_require__(/*! ./vitpose/image_processing_vitpose.js */ "./src/models/vitpose/image_processing_vitpose.js");
+/* harmony import */ var _yolos_image_processing_yolos_js__WEBPACK_IMPORTED_MODULE_37__ = __webpack_require__(/*! ./yolos/image_processing_yolos.js */ "./src/models/yolos/image_processing_yolos.js");
+
 
 
 
@@ -17793,6 +18031,7 @@ class VLMImageProcessor extends _base_image_processors_utils_js__WEBPACK_IMPORTE
             },
             ...config,
         });
+        // @ts-expect-error TS2339
         this.constant_values = this.config.background_color.map(x => x * this.rescale_factor)
     }
 
@@ -18241,6 +18480,8 @@ class MgpstrProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODULE
      * - bpe_preds: The list of BPE decoded sentences.
      * - wp_preds: The list of wp decoded sentences.
      */
+    // @ts-expect-error The type of this method is not compatible with the one
+    // in the base class. It might be a good idea to fix this.
     batch_decode([char_logits, bpe_logits, wp_logits]) {
         const [char_preds, char_scores] = this._decode_helper(char_logits, 'char');
         const [bpe_preds, bpe_scores] = this._decode_helper(bpe_logits, 'bpe');
@@ -18634,6 +18875,7 @@ class PaliGemmaProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MOD
         }
 
         const bos_token = this.tokenizer.bos_token;
+        // @ts-expect-error TS2339
         const image_seq_length = this.image_processor.config.image_seq_length;
         let input_strings;
         if (text.some((t) => t.includes(IMAGE_TOKEN))) {
@@ -18886,7 +19128,7 @@ class Phi3VProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODULE_
      * 
      * @param {string|string[]} text 
      * @param {RawImage|RawImage[]} images 
-     * @param  {...any} args 
+     * @param  { { padding?: boolean, truncation?: boolean, num_crops?: number } | undefined } options
      * @returns {Promise<any>}
      */
     async _call(text, images = null, {
@@ -18937,36 +19179,39 @@ class Phi3VProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODULE_
 __webpack_require__.r(__webpack_exports__);
 /* harmony export */ __webpack_require__.d(__webpack_exports__, {
 /* harmony export */   Florence2Processor: () => (/* reexport safe */ _florence2_processing_florence2_js__WEBPACK_IMPORTED_MODULE_0__.Florence2Processor),
-/* harmony export */   Idefics3Processor: () => (/* reexport safe */ _idefics3_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_3__.Idefics3Processor),
-/* harmony export */   JinaCLIPProcessor: () => (/* reexport safe */ _jina_clip_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_5__.JinaCLIPProcessor),
-/* harmony export */   MgpstrProcessor: () => (/* reexport safe */ _mgp_str_processing_mgp_str_js__WEBPACK_IMPORTED_MODULE_1__.MgpstrProcessor),
-/* harmony export */   MoonshineProcessor: () => (/* reexport safe */ _moonshine_processing_moonshine_js__WEBPACK_IMPORTED_MODULE_2__.MoonshineProcessor),
-/* harmony export */   OwlViTProcessor: () => (/* reexport safe */ _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_6__.OwlViTProcessor),
-/* harmony export */   PaliGemmaProcessor: () => (/* reexport safe */ _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_8__.PaliGemmaProcessor),
-/* harmony export */   Phi3VProcessor: () => (/* reexport safe */ _phi3_v_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_7__.Phi3VProcessor),
-/* harmony export */   PyAnnoteProcessor: () => (/* reexport safe */ _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_9__.PyAnnoteProcessor),
-/* harmony export */   Qwen2VLProcessor: () => (/* reexport safe */ _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_10__.Qwen2VLProcessor),
-/* harmony export */   SamProcessor: () => (/* reexport safe */ _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_11__.SamProcessor),
-/* harmony export */   SpeechT5Processor: () => (/* reexport safe */ _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_12__.SpeechT5Processor),
-/* harmony export */   VLChatProcessor: () => (/* reexport safe */ _janus_processing_janus_js__WEBPACK_IMPORTED_MODULE_4__.VLChatProcessor),
-/* harmony export */   Wav2Vec2ProcessorWithLM: () => (/* reexport safe */ _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_13__.Wav2Vec2ProcessorWithLM),
-/* harmony export */   WhisperProcessor: () => (/* reexport safe */ _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_14__.WhisperProcessor)
+/* harmony export */   GroundingDinoProcessor: () => (/* reexport safe */ _grounding_dino_processing_grounding_dino_js__WEBPACK_IMPORTED_MODULE_1__.GroundingDinoProcessor),
+/* harmony export */   Idefics3Processor: () => (/* reexport safe */ _idefics3_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_2__.Idefics3Processor),
+/* harmony export */   JinaCLIPProcessor: () => (/* reexport safe */ _jina_clip_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_4__.JinaCLIPProcessor),
+/* harmony export */   MgpstrProcessor: () => (/* reexport safe */ _mgp_str_processing_mgp_str_js__WEBPACK_IMPORTED_MODULE_5__.MgpstrProcessor),
+/* harmony export */   MoonshineProcessor: () => (/* reexport safe */ _moonshine_processing_moonshine_js__WEBPACK_IMPORTED_MODULE_6__.MoonshineProcessor),
+/* harmony export */   OwlViTProcessor: () => (/* reexport safe */ _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_7__.OwlViTProcessor),
+/* harmony export */   PaliGemmaProcessor: () => (/* reexport safe */ _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_9__.PaliGemmaProcessor),
+/* harmony export */   Phi3VProcessor: () => (/* reexport safe */ _phi3_v_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_8__.Phi3VProcessor),
+/* harmony export */   PyAnnoteProcessor: () => (/* reexport safe */ _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_10__.PyAnnoteProcessor),
+/* harmony export */   Qwen2VLProcessor: () => (/* reexport safe */ _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_11__.Qwen2VLProcessor),
+/* harmony export */   SamProcessor: () => (/* reexport safe */ _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_12__.SamProcessor),
+/* harmony export */   SpeechT5Processor: () => (/* reexport safe */ _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_13__.SpeechT5Processor),
+/* harmony export */   VLChatProcessor: () => (/* reexport safe */ _janus_processing_janus_js__WEBPACK_IMPORTED_MODULE_3__.VLChatProcessor),
+/* harmony export */   Wav2Vec2ProcessorWithLM: () => (/* reexport safe */ _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_14__.Wav2Vec2ProcessorWithLM),
+/* harmony export */   WhisperProcessor: () => (/* reexport safe */ _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_15__.WhisperProcessor)
 /* harmony export */ });
 /* harmony import */ var _florence2_processing_florence2_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ./florence2/processing_florence2.js */ "./src/models/florence2/processing_florence2.js");
-/* harmony import */ var _mgp_str_processing_mgp_str_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ./mgp_str/processing_mgp_str.js */ "./src/models/mgp_str/processing_mgp_str.js");
-/* harmony import */ var _moonshine_processing_moonshine_js__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! ./moonshine/processing_moonshine.js */ "./src/models/moonshine/processing_moonshine.js");
-/* harmony import */ var _idefics3_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! ./idefics3/processing_idefics3.js */ "./src/models/idefics3/processing_idefics3.js");
-/* harmony import */ var _janus_processing_janus_js__WEBPACK_IMPORTED_MODULE_4__ = __webpack_require__(/*! ./janus/processing_janus.js */ "./src/models/janus/processing_janus.js");
-/* harmony import */ var _jina_clip_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_5__ = __webpack_require__(/*! ./jina_clip/processing_jina_clip.js */ "./src/models/jina_clip/processing_jina_clip.js");
-/* harmony import */ var _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_6__ = __webpack_require__(/*! ./owlvit/processing_owlvit.js */ "./src/models/owlvit/processing_owlvit.js");
-/* harmony import */ var _phi3_v_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_7__ = __webpack_require__(/*! ./phi3_v/processing_phi3_v.js */ "./src/models/phi3_v/processing_phi3_v.js");
-/* harmony import */ var _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_8__ = __webpack_require__(/*! ./paligemma/processing_paligemma.js */ "./src/models/paligemma/processing_paligemma.js");
-/* harmony import */ var _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_9__ = __webpack_require__(/*! ./pyannote/processing_pyannote.js */ "./src/models/pyannote/processing_pyannote.js");
-/* harmony import */ var _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_10__ = __webpack_require__(/*! ./qwen2_vl/processing_qwen2_vl.js */ "./src/models/qwen2_vl/processing_qwen2_vl.js");
-/* harmony import */ var _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./sam/processing_sam.js */ "./src/models/sam/processing_sam.js");
-/* harmony import */ var _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_12__ = __webpack_require__(/*! ./speecht5/processing_speecht5.js */ "./src/models/speecht5/processing_speecht5.js");
-/* harmony import */ var _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_13__ = __webpack_require__(/*! ./wav2vec2/processing_wav2vec2.js */ "./src/models/wav2vec2/processing_wav2vec2.js");
-/* harmony import */ var _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_14__ = __webpack_require__(/*! ./whisper/processing_whisper.js */ "./src/models/whisper/processing_whisper.js");
+/* harmony import */ var _grounding_dino_processing_grounding_dino_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ./grounding_dino/processing_grounding_dino.js */ "./src/models/grounding_dino/processing_grounding_dino.js");
+/* harmony import */ var _idefics3_processing_idefics3_js__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! ./idefics3/processing_idefics3.js */ "./src/models/idefics3/processing_idefics3.js");
+/* harmony import */ var _janus_processing_janus_js__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! ./janus/processing_janus.js */ "./src/models/janus/processing_janus.js");
+/* harmony import */ var _jina_clip_processing_jina_clip_js__WEBPACK_IMPORTED_MODULE_4__ = __webpack_require__(/*! ./jina_clip/processing_jina_clip.js */ "./src/models/jina_clip/processing_jina_clip.js");
+/* harmony import */ var _mgp_str_processing_mgp_str_js__WEBPACK_IMPORTED_MODULE_5__ = __webpack_require__(/*! ./mgp_str/processing_mgp_str.js */ "./src/models/mgp_str/processing_mgp_str.js");
+/* harmony import */ var _moonshine_processing_moonshine_js__WEBPACK_IMPORTED_MODULE_6__ = __webpack_require__(/*! ./moonshine/processing_moonshine.js */ "./src/models/moonshine/processing_moonshine.js");
+/* harmony import */ var _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_7__ = __webpack_require__(/*! ./owlvit/processing_owlvit.js */ "./src/models/owlvit/processing_owlvit.js");
+/* harmony import */ var _phi3_v_processing_phi3_v_js__WEBPACK_IMPORTED_MODULE_8__ = __webpack_require__(/*! ./phi3_v/processing_phi3_v.js */ "./src/models/phi3_v/processing_phi3_v.js");
+/* harmony import */ var _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_9__ = __webpack_require__(/*! ./paligemma/processing_paligemma.js */ "./src/models/paligemma/processing_paligemma.js");
+/* harmony import */ var _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_10__ = __webpack_require__(/*! ./pyannote/processing_pyannote.js */ "./src/models/pyannote/processing_pyannote.js");
+/* harmony import */ var _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./qwen2_vl/processing_qwen2_vl.js */ "./src/models/qwen2_vl/processing_qwen2_vl.js");
+/* harmony import */ var _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_12__ = __webpack_require__(/*! ./sam/processing_sam.js */ "./src/models/sam/processing_sam.js");
+/* harmony import */ var _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_13__ = __webpack_require__(/*! ./speecht5/processing_speecht5.js */ "./src/models/speecht5/processing_speecht5.js");
+/* harmony import */ var _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_14__ = __webpack_require__(/*! ./wav2vec2/processing_wav2vec2.js */ "./src/models/wav2vec2/processing_wav2vec2.js");
+/* harmony import */ var _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_15__ = __webpack_require__(/*! ./whisper/processing_whisper.js */ "./src/models/whisper/processing_whisper.js");
+
 
 
 
@@ -19073,6 +19318,7 @@ class PyAnnoteFeatureExtractor extends _base_feature_extraction_utils_js__WEBPAC
 
             let current_speaker = -1;
             for (let i = 0; i < scores.length; ++i) {
+                /** @type {number[]} */
                 const probabilities = (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_2__.softmax)(scores[i]);
                 const [score, id] = (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_2__.max)(probabilities);
                 const [start, end] = [i, i + 1];
@@ -19260,6 +19506,7 @@ class Qwen2VLProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODUL
         }
 
         if (image_grid_thw) {
+            // @ts-expect-error TS2551
             let merge_length = this.image_processor.config.merge_size ** 2;
             let index = 0;
 
@@ -19751,8 +19998,8 @@ class SeamlessM4TFeatureExtractor extends _base_feature_extraction_utils_js__WEB
                         'int64',
                         new BigInt64Array(numPaddedFrames),
                         [1, numPaddedFrames],
-                    )
-                    padded_attention_mask.data.fill(1n, 0, num_frames);
+                    );
+                    /** @type {BigInt64Array} */ (padded_attention_mask.data).fill(1n, 0, num_frames);
                 }
             }
         }
@@ -20565,7 +20812,7 @@ class WhisperFeatureExtractor extends _base_feature_extraction_utils_js__WEBPACK
         )
 
         const data = features.data;
-        const maxValue = (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_3__.max)(data)[0];
+        const maxValue = (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_3__.max)(/** @type {Float32Array} */(data))[0];
 
         for (let i = 0; i < data.length; ++i) {
             data[i] = (Math.max(data[i], maxValue - 8.0) + 4.0) / 4.0;
@@ -20828,6 +21075,16 @@ class TensorOpRegistry {
         // executionProviders: ['webgpu'],
     };
 
+    static get nearest_interpolate_4d() {
+        if (!this._nearest_interpolate_4d) {
+            this._nearest_interpolate_4d = wrap(
+                [8, 10, 18, 0, 58, 129, 1, 10, 41, 10, 1, 120, 10, 0, 10, 0, 10, 1, 115, 18, 1, 121, 34, 6, 82, 101, 115, 105, 122, 101, 42, 18, 10, 4, 109, 111, 100, 101, 34, 7, 110, 101, 97, 114, 101, 115, 116, 160, 1, 3, 18, 1, 114, 90, 31, 10, 1, 120, 18, 26, 10, 24, 8, 1, 18, 20, 10, 3, 18, 1, 98, 10, 3, 18, 1, 99, 10, 3, 18, 1, 104, 10, 3, 18, 1, 119, 90, 15, 10, 1, 115, 18, 10, 10, 8, 8, 7, 18, 4, 10, 2, 8, 4, 98, 31, 10, 1, 121, 18, 26, 10, 24, 8, 1, 18, 20, 10, 3, 18, 1, 98, 10, 3, 18, 1, 99, 10, 3, 18, 1, 104, 10, 3, 18, 1, 119, 66, 2, 16, 21],
+                this.session_options,
+                'y',
+            );
+        }
+        return this._nearest_interpolate_4d;
+    }
     static get bilinear_interpolate_4d() {
         if (!this._bilinear_interpolate_4d) {
             this._bilinear_interpolate_4d = wrap(
@@ -21202,6 +21459,7 @@ class TextClassificationPipeline extends (/** @type {new (options: TextPipelineC
 
         // TODO: Use softmax tensor function
         const function_to_apply =
+            // @ts-expect-error TS2339
             this.model.config.problem_type === 'multi_label_classification'
                 ? batch => batch.sigmoid()
                 : batch => new _utils_tensor_js__WEBPACK_IMPORTED_MODULE_8__.Tensor(
@@ -21210,6 +21468,7 @@ class TextClassificationPipeline extends (/** @type {new (options: TextPipelineC
                     batch.dims,
                 ); // single_label_classification (default)
 
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         const toReturn = [];
@@ -21312,6 +21571,7 @@ class TokenClassificationPipeline extends (/** @type {new (options: TextPipeline
         const outputs = await this.model(model_inputs)
 
         const logits = outputs.logits;
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         const toReturn = [];
@@ -21651,11 +21911,14 @@ class Text2TextGenerationPipeline extends (/** @type {new (options: TextPipeline
 
 
         // Add global prefix, if present
+        // @ts-expect-error TS2339
         if (this.model.config.prefix) {
+            // @ts-expect-error TS2339
             texts = texts.map(x => this.model.config.prefix + x)
         }
 
         // Handle task specific params:
+        // @ts-expect-error TS2339
         const task_specific_params = this.model.config.task_specific_params
         if (task_specific_params && task_specific_params[this.task]) {
             // Add prefixes, if present
@@ -22394,6 +22657,7 @@ class AudioClassificationPipeline extends (/** @type {new (options: AudioPipelin
         const sampling_rate = this.processor.feature_extractor.config.sampling_rate;
         const preparedAudios = await prepareAudios(audio, sampling_rate);
 
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         const toReturn = [];
@@ -22704,6 +22968,7 @@ class AutomaticSpeechRecognitionPipeline extends (/** @type {new (options: TextA
             audio = [/** @type {AudioInput} */ (audio)];
         }
 
+        // @ts-expect-error TS2339
         const time_precision = this.processor.feature_extractor.config.chunk_length / this.model.config.max_source_positions;
         const hop_length = this.processor.feature_extractor.config.hop_length;
 
@@ -22769,7 +23034,9 @@ class AutomaticSpeechRecognitionPipeline extends (/** @type {new (options: TextA
 
                 // TODO: Right now we only get top beam
                 if (return_timestamps === 'word') {
+                    // @ts-expect-error TS2339
                     chunk.tokens = data.sequences.tolist()[0];
+                    // @ts-expect-error TS2339
                     chunk.token_timestamps = data.token_timestamps.tolist()[0].map(
                         (/** @type {number} */ x) => (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_6__.round)(x, 2)
                     );
@@ -22814,7 +23081,7 @@ class AutomaticSpeechRecognitionPipeline extends (/** @type {new (options: TextA
             const max_new_tokens = Math.floor(aud.length / sampling_rate) * 6;
             const outputs = await this.model.generate({ max_new_tokens, ...kwargs, ...inputs });
 
-            const text = this.processor.batch_decode(outputs, { skip_special_tokens: true })[0];
+            const text = this.processor.batch_decode(/** @type {Tensor} */(outputs), { skip_special_tokens: true })[0];
             toReturn.push({ text });
         }
         return single ? toReturn[0] : toReturn;
@@ -22963,6 +23230,7 @@ class ImageClassificationPipeline extends (/** @type {new (options: ImagePipelin
         const { pixel_values } = await this.processor(preparedImages);
         const output = await this.model({ pixel_values });
 
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         /** @type {ImageClassificationOutput[]} */
@@ -23077,6 +23345,7 @@ class ImageSegmentationPipeline extends (/** @type {new (options: ImagePipelineC
             }
         }
 
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         /** @type {ImageSegmentationPipelineOutput[]} */
@@ -23303,6 +23572,7 @@ class ObjectDetectionPipeline extends (/** @type {new (options: ImagePipelineCon
         const processed = this.processor.image_processor.post_process_object_detection(output, threshold, imageSizes);
 
         // Add labels
+        // @ts-expect-error TS2339
         const id2label = this.model.config.id2label;
 
         // Format output
@@ -23447,13 +23717,35 @@ class ZeroShotObjectDetectionPipeline extends (/** @type {new (options: TextImag
             // Run model with both text and pixel inputs
             const output = await this.model({ ...text_inputs, pixel_values });
 
-            // @ts-ignore
-            const processed = this.processor.image_processor.post_process_object_detection(output, threshold, imageSize, true)[0];
-            let result = processed.boxes.map((box, i) => ({
-                score: processed.scores[i],
-                label: candidate_labels[processed.classes[i]],
-                box: get_bounding_box(box, !percentage),
-            })).sort((a, b) => b.score - a.score);
+            let result;
+            if('post_process_grounded_object_detection' in this.processor) {
+                // @ts-ignore
+                const processed = this.processor.post_process_grounded_object_detection(
+                    output,
+                    text_inputs.input_ids,
+                    {
+                        // TODO: support separate threshold values
+                        box_threshold: threshold,
+                        text_threshold: threshold,
+                        target_sizes: imageSize,
+                    },
+                )[0];
+                result = processed.boxes.map((box, i) => ({
+                    score: processed.scores[i],
+                    label: processed.labels[i],
+                    box: get_bounding_box(box, !percentage),
+                }))
+            } else {
+                // @ts-ignore
+                const processed = this.processor.image_processor.post_process_object_detection(output, threshold, imageSize, true)[0];
+                result = processed.boxes.map((box, i) => ({
+                    score: processed.scores[i],
+                    label: candidate_labels[processed.classes[i]],
+                    box: get_bounding_box(box, !percentage),
+                }))
+            }
+            result.sort((a, b) => b.score - a.score);
+
             if (top_k !== null) {
                 result = result.slice(0, top_k);
             }
@@ -23522,6 +23814,7 @@ class DocumentQuestionAnsweringPipeline extends (/** @type {new (options: TextIm
         // Run model
         const output = await this.model.generate({
             inputs: pixel_values,
+            // @ts-expect-error TS2339
             max_length: this.model.config.decoder.max_position_embeddings,
             decoder_input_ids,
             ...generate_kwargs,
@@ -23572,7 +23865,7 @@ class DocumentQuestionAnsweringPipeline extends (/** @type {new (options: TextIm
  * const synthesizer = await pipeline('text-to-speech', 'Xenova/speecht5_tts', { quantized: false });
  * const speaker_embeddings = 'https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/speaker_embeddings.bin';
  * const out = await synthesizer('Hello, my dog is cute', { speaker_embeddings });
- * // {
+ * // RawAudio {
  * //   audio: Float32Array(26112) [-0.00005657337896991521, 0.00020583874720614403, ...],
  * //   sampling_rate: 16000
  * // }
@@ -23592,7 +23885,7 @@ class DocumentQuestionAnsweringPipeline extends (/** @type {new (options: TextIm
  * ```javascript
  * const synthesizer = await pipeline('text-to-speech', 'Xenova/mms-tts-fra');
  * const out = await synthesizer('Bonjour');
- * // {
+ * // RawAudio {
  * //   audio: Float32Array(23808) [-0.00037693005288019776, 0.0003325853613205254, ...],
  * //   sampling_rate: 16000
  * // }
@@ -23637,11 +23930,12 @@ class TextToAudioPipeline extends (/** @type {new (options: TextToAudioPipelineC
         // Generate waveform
         const { waveform } = await this.model(inputs);
 
+        // @ts-expect-error TS2339
         const sampling_rate = this.model.config.sampling_rate;
-        return {
-            audio: waveform.data,
+        return new _utils_audio_js__WEBPACK_IMPORTED_MODULE_7__.RawAudio(
+            waveform.data,
             sampling_rate,
-        }
+        )
     }
 
     async _call_text_to_spectrogram(text_inputs, { speaker_embeddings }) {
@@ -23681,10 +23975,10 @@ class TextToAudioPipeline extends (/** @type {new (options: TextToAudioPipelineC
         const { waveform } = await this.model.generate_speech(input_ids, speaker_embeddings, { vocoder: this.vocoder });
 
         const sampling_rate = this.processor.feature_extractor.config.sampling_rate;
-        return {
-            audio: waveform.data,
+        return new _utils_audio_js__WEBPACK_IMPORTED_MODULE_7__.RawAudio(
+            waveform.data,
             sampling_rate,
-        }
+        )
     }
 }
 
@@ -23794,11 +24088,23 @@ class DepthEstimationPipeline extends (/** @type {new (options: ImagePipelineCon
 
         const toReturn = [];
         for (let i = 0; i < preparedImages.length; ++i) {
-            const prediction = (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_8__.interpolate)(predicted_depth[i], preparedImages[i].size.reverse(), 'bilinear', false);
-            const formatted = prediction.mul_(255 / (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_6__.max)(prediction.data)[0]).to('uint8');
+            const batch = predicted_depth[i];
+            const [height, width] = batch.dims.slice(-2);
+            const [new_width, new_height] = preparedImages[i].size;
+
+            // Interpolate to original size
+            const prediction = (await (0,_utils_tensor_js__WEBPACK_IMPORTED_MODULE_8__.interpolate_4d)(batch.view(1, 1, height, width), {
+                size: [new_height, new_width],
+                mode: 'bilinear',
+            })).view(new_height, new_width);
+
+            const minval = /** @type {number} */(prediction.min().item());
+            const maxval = /** @type {number} */(prediction.max().item());
+            const formatted = prediction.sub(minval).div_(maxval - minval).mul_(255).to('uint8').unsqueeze(0);
+            const depth = _utils_image_js__WEBPACK_IMPORTED_MODULE_9__.RawImage.fromTensor(formatted);
             toReturn.push({
-                predicted_depth: predicted_depth[i],
-                depth: _utils_image_js__WEBPACK_IMPORTED_MODULE_9__.RawImage.fromTensor(formatted),
+                predicted_depth: prediction,
+                depth,
             });
         }
 
@@ -24278,6 +24584,7 @@ async function loadItems(mapping, model, pretrainedOptions) {
     return result;
 }
 
+
 /***/ }),
 
 /***/ "./src/tokenizers.js":
@@ -24347,7 +24654,6 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony import */ var _utils_data_structures_js__WEBPACK_IMPORTED_MODULE_5__ = __webpack_require__(/*! ./utils/data-structures.js */ "./src/utils/data-structures.js");
 /* harmony import */ var _huggingface_jinja__WEBPACK_IMPORTED_MODULE_6__ = __webpack_require__(/*! @huggingface/jinja */ "./node_modules/@huggingface/jinja/dist/index.js");
 /* harmony import */ var _models_whisper_common_whisper_js__WEBPACK_IMPORTED_MODULE_7__ = __webpack_require__(/*! ./models/whisper/common_whisper.js */ "./src/models/whisper/common_whisper.js");
-/* harmony import */ var _utils_constants_js__WEBPACK_IMPORTED_MODULE_8__ = __webpack_require__(/*! ./utils/constants.js */ "./src/utils/constants.js");
 
 /**
  * @file Tokenizers are used to prepare textual inputs for a model.
@@ -24384,7 +24690,6 @@ __webpack_require__.r(__webpack_exports__);
 
 
 
-
 /**
  * @typedef {Object} TokenizerProperties Additional tokenizer-specific properties.
  * @property {boolean} [legacy=false] Whether or not the `legacy` behavior of the tokenizer should be used.
@@ -24868,7 +25173,7 @@ class Unigram extends TokenizerModel {
      * Create a new Unigram tokenizer model.
      * @param {Object} config The configuration object for the Unigram model.
      * @param {number} config.unk_id The ID of the unknown token
-     * @param {any[][]} config.vocab A 2D array representing a mapping of tokens to scores.
+     * @param {[string, number][]} config.vocab A 2D array representing a mapping of tokens to scores.
      * @param {Object} moreConfig Additional configuration object for the Unigram model.
      */
     constructor(config, moreConfig) {
@@ -24876,11 +25181,10 @@ class Unigram extends TokenizerModel {
 
         const vocabSize = config.vocab.length;
         this.vocab = new Array(vocabSize);
+        /** @type {number[]} */
         this.scores = new Array(vocabSize);
         for (let i = 0; i < vocabSize; ++i) {
-            const piece = config.vocab[i];
-            this.vocab[i] = piece[0];
-            this.scores[i] = piece[1];
+            [this.vocab[i], this.scores[i]] = config.vocab[i];
         }
 
         this.unk_token_id = config.unk_id;
@@ -28720,6 +29024,7 @@ class AutoTokenizer {
 "use strict";
 __webpack_require__.r(__webpack_exports__);
 /* harmony export */ __webpack_require__.d(__webpack_exports__, {
+/* harmony export */   RawAudio: () => (/* binding */ RawAudio),
 /* harmony export */   hamming: () => (/* binding */ hamming),
 /* harmony export */   hanning: () => (/* binding */ hanning),
 /* harmony export */   mel_filter_bank: () => (/* binding */ mel_filter_bank),
@@ -28730,7 +29035,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony import */ var _hub_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ./hub.js */ "./src/utils/hub.js");
 /* harmony import */ var _maths_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ./maths.js */ "./src/utils/maths.js");
 /* harmony import */ var _core_js__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! ./core.js */ "./src/utils/core.js");
-/* harmony import */ var _tensor_js__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! ./tensor.js */ "./src/utils/tensor.js");
+/* harmony import */ var _env_js__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! ../env.js */ "./src/env.js");
+/* harmony import */ var fs__WEBPACK_IMPORTED_MODULE_4__ = __webpack_require__(/*! fs */ "fs");
+/* harmony import */ var _tensor_js__WEBPACK_IMPORTED_MODULE_5__ = __webpack_require__(/*! ./tensor.js */ "./src/utils/tensor.js");
 /**
  * @file Helper module for audio processing. 
  * 
@@ -28746,6 +29053,8 @@ __webpack_require__.r(__webpack_exports__);
 
 
 
+
+
 /**
  * Helper function to read audio from a path/URL.
  * @param {string|URL} url The path/URL to load the audio from.
@@ -29339,10 +29648,10 @@ async function spectrogram(
     //  - mel_filters.shape=(80, 201)
     //  - magnitudes.shape=(3000, 201) => magnitudes.T.shape=(201, 3000)
     //  - mel_spec.shape=(80, 3000)
-    let mel_spec = await (0,_tensor_js__WEBPACK_IMPORTED_MODULE_3__.matmul)(
+    let mel_spec = await (0,_tensor_js__WEBPACK_IMPORTED_MODULE_5__.matmul)(
         // TODO: Make `mel_filters` a Tensor during initialization
-        new _tensor_js__WEBPACK_IMPORTED_MODULE_3__.Tensor('float32', mel_filters.flat(), [num_mel_filters, num_frequency_bins]),
-        new _tensor_js__WEBPACK_IMPORTED_MODULE_3__.Tensor('float32', transposedMagnitudeData, [num_frequency_bins, d1Max]),
+        new _tensor_js__WEBPACK_IMPORTED_MODULE_5__.Tensor('float32', mel_filters.flat(), [num_mel_filters, num_frequency_bins]),
+        new _tensor_js__WEBPACK_IMPORTED_MODULE_5__.Tensor('float32', transposedMagnitudeData, [num_frequency_bins, d1Max]),
     );
     if (transpose) {
         mel_spec = mel_spec.transpose(1, 0);
@@ -29432,6 +29741,116 @@ function window_function(window_length, name, {
     return window;
 }
 
+/**
+ * Encode audio data to a WAV file.
+ * WAV file specs : https://en.wikipedia.org/wiki/WAV#WAV_File_header
+ * 
+ * Adapted from https://www.npmjs.com/package/audiobuffer-to-wav
+ * @param {Float32Array} samples The audio samples.
+ * @param {number} rate The sample rate.
+ * @returns {ArrayBuffer} The WAV audio buffer.
+ */
+function encodeWAV(samples, rate) {
+    let offset = 44;
+    const buffer = new ArrayBuffer(offset + samples.length * 4);
+    const view = new DataView(buffer);
+
+    /* RIFF identifier */
+    writeString(view, 0, "RIFF");
+    /* RIFF chunk length */
+    view.setUint32(4, 36 + samples.length * 4, true);
+    /* RIFF type */
+    writeString(view, 8, "WAVE");
+    /* format chunk identifier */
+    writeString(view, 12, "fmt ");
+    /* format chunk length */
+    view.setUint32(16, 16, true);
+    /* sample format (raw) */
+    view.setUint16(20, 3, true);
+    /* channel count */
+    view.setUint16(22, 1, true);
+    /* sample rate */
+    view.setUint32(24, rate, true);
+    /* byte rate (sample rate * block align) */
+    view.setUint32(28, rate * 4, true);
+    /* block align (channel count * bytes per sample) */
+    view.setUint16(32, 4, true);
+    /* bits per sample */
+    view.setUint16(34, 32, true);
+    /* data chunk identifier */
+    writeString(view, 36, "data");
+    /* data chunk length */
+    view.setUint32(40, samples.length * 4, true);
+
+    for (let i = 0; i < samples.length; ++i, offset += 4) {
+        view.setFloat32(offset, samples[i], true);
+    }
+
+    return buffer;
+}
+
+function writeString(view, offset, string) {
+    for (let i = 0; i < string.length; ++i) {
+        view.setUint8(offset + i, string.charCodeAt(i));
+    }
+}
+
+
+class RawAudio {
+
+    /**
+     * Create a new `RawAudio` object.
+     * @param {Float32Array} audio Audio data
+     * @param {number} sampling_rate Sampling rate of the audio data
+     */
+    constructor(audio, sampling_rate) {
+        this.audio = audio
+        this.sampling_rate = sampling_rate
+    }
+
+    /**
+     * Convert the audio to a wav file buffer.
+     * @returns {ArrayBuffer} The WAV file.
+     */
+    toWav() {
+        return encodeWAV(this.audio, this.sampling_rate)
+    }
+
+    /**
+     * Convert the audio to a blob.
+     * @returns {Blob}
+     */
+    toBlob() {
+        const wav = this.toWav();
+        const blob = new Blob([wav], { type: 'audio/wav' });
+        return blob;
+    }
+
+    /**
+     * Save the audio to a wav file.
+     * @param {string} path
+     */
+    async save(path) {
+        let fn;
+
+        if (_env_js__WEBPACK_IMPORTED_MODULE_3__.apis.IS_BROWSER_ENV) {
+            if (_env_js__WEBPACK_IMPORTED_MODULE_3__.apis.IS_WEBWORKER_ENV) {
+                throw new Error('Unable to save a file from a Web Worker.')
+            }
+            fn = _core_js__WEBPACK_IMPORTED_MODULE_2__.saveBlob;
+        } else if (_env_js__WEBPACK_IMPORTED_MODULE_3__.apis.IS_FS_AVAILABLE) {
+            fn = async (/** @type {string} */ path, /** @type {Blob} */ blob) => {
+                let buffer = await blob.arrayBuffer();
+                fs__WEBPACK_IMPORTED_MODULE_4__.writeFileSync(path, Buffer.from(buffer));
+            }
+        } else {
+            throw new Error('Unable to save because filesystem is disabled in this environment.')
+        }
+
+        await fn(path, this.toBlob())
+    }
+}
+
 
 /***/ }),
 
@@ -29487,7 +29906,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   pick: () => (/* binding */ pick),
 /* harmony export */   pop: () => (/* binding */ pop),
 /* harmony export */   product: () => (/* binding */ product),
-/* harmony export */   reverseDictionary: () => (/* binding */ reverseDictionary)
+/* harmony export */   reverseDictionary: () => (/* binding */ reverseDictionary),
+/* harmony export */   saveBlob: () => (/* binding */ saveBlob)
 /* harmony export */ });
 
 /**
@@ -29680,6 +30100,32 @@ function calculateReflectOffset(i, w) {
     return Math.abs((i + w) % (2 * w) - w);
 }
 
+/**
+ * Save blob file on the web.
+ * @param {string} path The path to save the blob to
+ * @param {Blob} blob The blob to save
+ */
+function saveBlob(path, blob){
+    // Convert the canvas content to a data URL
+    const dataURL = URL.createObjectURL(blob);
+
+    // Create an anchor element with the data URL as the href attribute
+    const downloadLink = document.createElement('a');
+    downloadLink.href = dataURL;
+
+    // Set the download attribute to specify the desired filename for the downloaded image
+    downloadLink.download = path;
+
+    // Trigger the download
+    downloadLink.click();
+
+    // Clean up: remove the anchor element from the DOM
+    downloadLink.remove();
+
+    // Revoke the Object URL to free up memory
+    URL.revokeObjectURL(dataURL);
+}
+
 /**
  *
  * @param {Object} o
@@ -30243,6 +30689,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */ });
 /* harmony import */ var _env_js__WEBPACK_IMPORTED_MODULE_0__ = __webpack_require__(/*! ../env.js */ "./src/env.js");
 /* harmony import */ var _devices_js__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! ./devices.js */ "./src/utils/devices.js");
+/// <reference types="@webgpu/types" />
+
 
 
 
@@ -30498,7 +30946,7 @@ class FileResponse {
      */
     async arrayBuffer() {
         const data = await fs__WEBPACK_IMPORTED_MODULE_0__.promises.readFile(this.filePath);
-        return data.buffer;
+        return /** @type {ArrayBuffer} */ (data.buffer);
     }
 
     /**
@@ -31860,23 +32308,9 @@ class RawImage {
             // Convert image to Blob
             const blob = await this.toBlob(mime);
 
-            // Convert the canvas content to a data URL
-            const dataURL = URL.createObjectURL(blob);
+            (0,_core_js__WEBPACK_IMPORTED_MODULE_0__.saveBlob)(path, blob)
 
-            // Create an anchor element with the data URL as the href attribute
-            const downloadLink = document.createElement('a');
-            downloadLink.href = dataURL;
-
-            // Set the download attribute to specify the desired filename for the downloaded image
-            downloadLink.download = path;
-
-            // Trigger the download
-            downloadLink.click();
-
-            // Clean up: remove the anchor element from the DOM
-            downloadLink.remove();
-
-        } else if (!_env_js__WEBPACK_IMPORTED_MODULE_2__.env.useFS) {
+        } else if (!_env_js__WEBPACK_IMPORTED_MODULE_2__.apis.IS_FS_AVAILABLE) {
             throw new Error('Unable to save the image because filesystem is disabled in this environment.')
 
         } else {
@@ -31906,6 +32340,7 @@ class RawImage {
 const load_image = RawImage.read.bind(RawImage);
 
 
+
 /***/ }),
 
 /***/ "./src/utils/maths.js":
@@ -32159,8 +32594,9 @@ function magnitude(arr) {
 
 /**
  * Returns the value and index of the minimum element in an array.
- * @param {number[]|TypedArray} arr array of numbers.
- * @returns {[number, number]} the value and index of the minimum element, of the form: [valueOfMin, indexOfMin]
+ * @template {number[]|bigint[]|AnyTypedArray} T
+ * @param {T} arr array of numbers.
+ * @returns {T extends bigint[]|BigTypedArray ? [bigint, number] : [number, number]} the value and index of the minimum element, of the form: [valueOfMin, indexOfMin]
  * @throws {Error} If array is empty.
  */
 function min(arr) {
@@ -32173,14 +32609,15 @@ function min(arr) {
             indexOfMin = i;
         }
     }
-    return [min, indexOfMin];
+    return /** @type {T extends bigint[]|BigTypedArray ? [bigint, number] : [number, number]} */([min, indexOfMin]);
 }
 
 
 /**
  * Returns the value and index of the maximum element in an array.
- * @param {number[]|AnyTypedArray} arr array of numbers.
- * @returns {[number, number]} the value and index of the maximum element, of the form: [valueOfMax, indexOfMax]
+ * @template {number[]|bigint[]|AnyTypedArray} T
+ * @param {T} arr array of numbers.
+ * @returns {T extends bigint[]|BigTypedArray ? [bigint, number] : [number, number]} the value and index of the maximum element, of the form: [valueOfMax, indexOfMax]
  * @throws {Error} If array is empty.
  */
 function max(arr) {
@@ -32193,7 +32630,7 @@ function max(arr) {
             indexOfMax = i;
         }
     }
-    return [Number(max), indexOfMax];
+    return /** @type {T extends bigint[]|BigTypedArray ? [bigint, number] : [number, number]} */([max, indexOfMax]);
 }
 
 function isPowerOfTwo(number) {
@@ -33491,8 +33928,6 @@ class Tensor {
         return this.permute(...dims);
     }
 
-    // TODO add .max() and .min() methods
-
     /**
      * Returns the sum of each row of the input tensor in the given dimension dim.
      *
@@ -33521,55 +33956,22 @@ class Tensor {
         }
 
         const this_data = this.data;
+        const fn = (a, b) => a + (b ** p);
 
         if (dim === null) {
             // @ts-ignore
-            let val = this_data.reduce((a, b) => a + (b ** p), 0) ** (1 / p);
+            const val = this_data.reduce(fn, 0) ** (1 / p);
             return new Tensor(this.type, [val], []);
         }
 
-        // Negative indexing
-        dim = safeIndex(dim, this.dims.length);
-
-        // Calculate the shape of the resulting array after summation
-        const resultDims = this.dims.slice(); // Copy the original dimensions
-        resultDims[dim] = 1; // Remove the specified axis
-
-        // Create a new array to store the accumulated values
-        // @ts-ignore
-        const result = new this_data.constructor(this_data.length / this.dims[dim]);
-
-        // Iterate over the data array
-        for (let i = 0; i < this_data.length; ++i) {
-
-            // Calculate the index in the resulting array
-            let resultIndex = 0;
-
-            for (let j = this.dims.length - 1, num = i, resultMultiplier = 1; j >= 0; --j) {
-                const size = this.dims[j];
-                if (j !== dim) {
-                    const index = num % size;
-                    resultIndex += index * resultMultiplier;
-                    resultMultiplier *= resultDims[j];
-                }
-                num = Math.floor(num / size);
-            }
-
-            // Accumulate the value at the current index
-            result[resultIndex] += (this_data[i]) ** p;
-        }
+        const [type, result, resultDims] = reduce_helper(fn, this, dim, keepdim);
 
         if (p !== 1) {
             for (let i = 0; i < result.length; ++i) {
                 result[i] = result[i] ** (1 / p);
             }
         }
-
-        if (!keepdim) {
-            resultDims.splice(dim, 1);
-        }
-
-        return new Tensor(this.type, result, resultDims);
+        return new Tensor(type, result, resultDims);
     }
 
     /**
@@ -33632,7 +34034,7 @@ class Tensor {
      * NOTE: The returned tensor shares the storage with the input tensor, so changing the contents of one will change the contents of the other.
      * If you would like a copy, use `tensor.clone()` before squeezing.
      *
-     * @param {number} [dim=null] If given, the input will be squeezed only in the specified dimensions.
+     * @param {number|number[]} [dim=null] If given, the input will be squeezed only in the specified dimensions.
      * @returns {Tensor} The squeezed tensor
      */
     squeeze(dim = null) {
@@ -33742,6 +34144,34 @@ class Tensor {
         return this.clone().neg_();
     }
 
+    /**
+     * Computes input > val element-wise.
+     * @param {number} val The value to compare with.
+     * @returns {Tensor} A boolean tensor that is `true` where input is greater than other and `false` elsewhere.
+     */
+    gt(val) {
+        const mask = new Uint8Array(this.data.length);
+        const this_data = this.data;
+        for (let i = 0; i < this_data.length; ++i) {
+            mask[i] = this_data[i] > val ? 1 : 0;
+        }
+        return new Tensor('bool', mask, this.dims);
+    }
+
+    /**
+     * Computes input < val element-wise.
+     * @param {number} val The value to compare with.
+     * @returns {Tensor} A boolean tensor that is `true` where input is less than other and `false` elsewhere.
+     */
+    lt(val) {
+        const mask = new Uint8Array(this.data.length);
+        const this_data = this.data;
+        for (let i = 0; i < this_data.length; ++i) {
+            mask[i] = this_data[i] < val ? 1 : 0;
+        }
+        return new Tensor('bool', mask, this.dims);
+    }
+
     /**
      * In-place version of @see {@link Tensor.clamp}
      */
@@ -33786,6 +34216,41 @@ class Tensor {
         return mean(this, dim, keepdim);
     }
 
+    min(dim = null, keepdim = false) {
+        if (dim === null) {
+            // None to reduce over all dimensions.
+            const val = (0,_maths_js__WEBPACK_IMPORTED_MODULE_0__.min)(this.data)[0];
+            return new Tensor(this.type, [val], [/* scalar */]);
+        }
+        const [type, result, resultDims] = reduce_helper((a, b) => Math.min(a, b), this, dim, keepdim, Infinity);
+        return new Tensor(type, result, resultDims);
+    }
+
+    max(dim = null, keepdim = false) {
+        if (dim === null) {
+            // None to reduce over all dimensions.
+            const val = (0,_maths_js__WEBPACK_IMPORTED_MODULE_0__.max)(this.data)[0];
+            return new Tensor(this.type, [val], [/* scalar */]);
+        }
+        const [type, result, resultDims] = reduce_helper((a, b) => Math.max(a, b), this, dim, keepdim, -Infinity);
+        return new Tensor(type, result, resultDims);
+    }
+
+    argmin(dim = null, keepdim = false) {
+        if (dim !== null) {
+            throw new Error("`dim !== null` not yet implemented.");
+        }
+        const index = (0,_maths_js__WEBPACK_IMPORTED_MODULE_0__.min)(this.data)[1];
+        return new Tensor('int64', [BigInt(index)], []);
+    }
+    argmax(dim = null, keepdim = false) {
+        if (dim !== null) {
+            throw new Error("`dim !== null` not yet implemented.");
+        }
+        const index = (0,_maths_js__WEBPACK_IMPORTED_MODULE_0__.max)(this.data)[1];
+        return new Tensor('int64', [BigInt(index)], []);
+    }
+
     /**
      * Performs Tensor dtype conversion.
      * @param {DataType} type The desired data type.
@@ -33919,7 +34384,7 @@ function interpolate(input, [out_height, out_width], mode = 'bilinear', align_co
  * @param {Tensor} input the input tensor
  * @param {Object} options the options for the interpolation
  * @param {[number, number]|[number, number, number]|[number, number, number, number]} [options.size=null] output spatial size.
- * @param {"bilinear"|"bicubic"} [options.mode='bilinear'] algorithm used for upsampling
+ * @param {"nearest"|"bilinear"|"bicubic"} [options.mode='bilinear'] algorithm used for upsampling
  * @returns {Promise<Tensor>} The interpolated tensor.
  */
 async function interpolate_4d(input, {
@@ -33949,7 +34414,9 @@ async function interpolate_4d(input, {
     }
 
     let op;
-    if (mode === 'bilinear') {
+    if (mode === 'nearest') {
+        op = await _ops_registry_js__WEBPACK_IMPORTED_MODULE_2__.TensorOpRegistry.nearest_interpolate_4d;
+    } else if (mode === 'bilinear') {
         op = await _ops_registry_js__WEBPACK_IMPORTED_MODULE_2__.TensorOpRegistry.bilinear_interpolate_4d;
     } else if (mode === 'bicubic') {
         op = await _ops_registry_js__WEBPACK_IMPORTED_MODULE_2__.TensorOpRegistry.bicubic_interpolate_4d;
@@ -33990,13 +34457,13 @@ async function rfft(x, a) {
  * Returns the k largest elements of the given input tensor.
  * Inspired by https://pytorch.org/docs/stable/generated/torch.topk.html
  * @param {Tensor} x the input tensor
- * @param {number} k the k in "top-k"
+ * @param {number} [k] the k in "top-k"
  * @returns {Promise<[Tensor, Tensor]>} the output tuple of (Tensor, LongTensor) of top-k elements and their indices.
  */
 async function topk(x, k) {
     const op = await _ops_registry_js__WEBPACK_IMPORTED_MODULE_2__.TensorOpRegistry.top_k;
 
-    if (k === null) {
+    if (k == null) {
         k = x.dims.at(-1);
     } else {
         k = Math.min(k, x.dims.at(-1));
@@ -34025,10 +34492,10 @@ const arrayToIndexTensor = (array) => new Tensor('int64', array, [array.length])
 async function slice(data, starts, ends, axes, steps) {
     const op = await _ops_registry_js__WEBPACK_IMPORTED_MODULE_2__.TensorOpRegistry.slice;
     return await op({
-        x: data, 
-        s: arrayToIndexTensor(starts), 
-        e: arrayToIndexTensor(ends), 
-        a: arrayToIndexTensor(axes), 
+        x: data,
+        s: arrayToIndexTensor(starts),
+        e: arrayToIndexTensor(ends),
+        a: arrayToIndexTensor(axes),
         t: arrayToIndexTensor(steps ?? new Array(axes.length).fill(1)),
     });
 }
@@ -34263,35 +34730,19 @@ function stack(tensors, dim = 0) {
 
 
 /**
- * Calculates the standard deviation and mean over the dimensions specified by dim. dim can be a single dimension or `null` to reduce over all dimensions.
- * @param {Tensor} input the input tenso
- * @param {number|null} dim the dimension to reduce. If None, all dimensions are reduced.
- * @param {number} correction difference between the sample size and sample degrees of freedom. Defaults to Bessel's correction, correction=1.
+ * @param {(previousValue: any, currentValue: any, currentIndex?: number, resultIndex?: number) => any} callbackfn
+ * @param {Tensor} input the input tensor.
+ * @param {number|null} dim the dimension to reduce.
  * @param {boolean} keepdim whether the output tensor has dim retained or not.
- * @returns {Tensor[]} A tuple of (std, mean) tensors.
+ * @returns {[DataType, any, number[]]} The reduced tensor data.
  */
-function std_mean(input, dim = null, correction = 1, keepdim = false) {
-    const inputData = /** @type {Float32Array} */(input.data);
+function reduce_helper(callbackfn, input, dim = null, keepdim = false, initialValue = null) {
+    const inputData = input.data;
     const inputDims = input.dims;
 
-    if (dim === null) {
-        // None to reduce over all dimensions.
-        const sum = inputData.reduce((a, b) => a + b, 0);
-        const mean = sum / inputData.length;
-        const std = Math.sqrt(inputData.reduce((a, b) => a + (b - mean) ** 2, 0) / (inputData.length - correction));
-
-        const meanTensor = new Tensor(input.type, [mean], [/* scalar */]);
-        const stdTensor = new Tensor(input.type, [std], [/* scalar */]);
-
-        return [stdTensor, meanTensor];
-    }
-
     // Negative indexing
     dim = safeIndex(dim, inputDims.length);
 
-    const meanTensor = mean(input, dim, keepdim);
-    const meanTensorData = meanTensor.data;
-
     // Calculate the shape of the resulting array after summation
     const resultDims = inputDims.slice(); // Copy the original dimensions
     resultDims[dim] = 1; // Remove the specified axis
@@ -34299,6 +34750,9 @@ function std_mean(input, dim = null, correction = 1, keepdim = false) {
     // Create a new array to store the accumulated values
     // @ts-ignore
     const result = new inputData.constructor(inputData.length / inputDims[dim]);
+    if (initialValue !== null) {
+        result.fill(initialValue);
+    }
 
     // Iterate over the data array
     for (let i = 0; i < inputData.length; ++i) {
@@ -34317,23 +34771,55 @@ function std_mean(input, dim = null, correction = 1, keepdim = false) {
         }
 
         // Accumulate the value at the current index
-        result[resultIndex] += (inputData[i] - meanTensorData[resultIndex]) ** 2;
+        result[resultIndex] = callbackfn(result[resultIndex], inputData[i], i, resultIndex);
     }
 
-    for (let i = 0; i < result.length; ++i) {
-        result[i] = Math.sqrt(result[i] / (inputDims[dim] - correction));
+    if (!keepdim) resultDims.splice(dim, 1);
+
+    return [input.type, result, resultDims];
+}
+
+
+/**
+ * Calculates the standard deviation and mean over the dimensions specified by dim. dim can be a single dimension or `null` to reduce over all dimensions.
+ * @param {Tensor} input the input tenso
+ * @param {number|null} dim the dimension to reduce. If None, all dimensions are reduced.
+ * @param {number} correction difference between the sample size and sample degrees of freedom. Defaults to Bessel's correction, correction=1.
+ * @param {boolean} keepdim whether the output tensor has dim retained or not.
+ * @returns {Tensor[]} A tuple of (std, mean) tensors.
+ */
+function std_mean(input, dim = null, correction = 1, keepdim = false) {
+    const inputData = /** @type {Float32Array} */(input.data);
+    const inputDims = input.dims;
+
+    if (dim === null) {
+        // None to reduce over all dimensions.
+        const sum = inputData.reduce((a, b) => a + b, 0);
+        const mean = sum / inputData.length;
+        const std = Math.sqrt(inputData.reduce((a, b) => a + (b - mean) ** 2, 0) / (inputData.length - correction));
+
+        const meanTensor = new Tensor(input.type, [mean], [/* scalar */]);
+        const stdTensor = new Tensor(input.type, [std], [/* scalar */]);
+
+        return [stdTensor, meanTensor];
     }
+    dim = safeIndex(dim, inputDims.length);
+    const meanTensor = mean(input, dim, keepdim);
+    const meanTensorData = meanTensor.data;
 
-    if (!keepdim) {
-        resultDims.splice(dim, 1);
+    // Compute squared sum
+    const [type, result, resultDims] = reduce_helper((a, b, i, j) => a + (b - meanTensorData[j]) ** 2, input, dim, keepdim);
+
+    // Square root of the squared sum
+    for (let i = 0; i < result.length; ++i) {
+        result[i] = Math.sqrt(result[i] / (inputDims[dim] - correction));
     }
 
-    const stdTensor = new Tensor(input.type, result, resultDims);
+    const stdTensor = new Tensor(type, result, resultDims);
 
     return [stdTensor, meanTensor];
 }
 
-
 /**
  * Returns the mean value of each row of the input tensor in the given dimension dim.
  * @param {Tensor} input the input tensor.
@@ -34342,58 +34828,27 @@ function std_mean(input, dim = null, correction = 1, keepdim = false) {
  * @returns {Tensor} A new tensor with means taken along the specified dimension.
  */
 function mean(input, dim = null, keepdim = false) {
+    const inputDims = input.dims;
     const inputData = /** @type {Float32Array} */(input.data);
 
     if (dim === null) {
         // None to reduce over all dimensions.
-        // @ts-ignore
         const val = inputData.reduce((a, b) => a + b, 0);
         return new Tensor(input.type, [val / inputData.length], [/* scalar */]);
     }
-    const inputDims = input.dims;
-
-    // Negative indexing
     dim = safeIndex(dim, inputDims.length);
 
-    // Calculate the shape of the resulting array after summation
-    const resultDims = inputDims.slice(); // Copy the original dimensions
-    resultDims[dim] = 1; // Remove the specified axis
-
-    // Create a new array to store the accumulated values
-    // @ts-ignore
-    const result = new inputData.constructor(inputData.length / inputDims[dim]);
-
-    // Iterate over the data array
-    for (let i = 0; i < inputData.length; ++i) {
-
-        // Calculate the index in the resulting array
-        let resultIndex = 0;
-
-        for (let j = inputDims.length - 1, num = i, resultMultiplier = 1; j >= 0; --j) {
-            const size = inputDims[j];
-            if (j !== dim) {
-                const index = num % size;
-                resultIndex += index * resultMultiplier;
-                resultMultiplier *= resultDims[j];
-            }
-            num = Math.floor(num / size);
-        }
-
-        // Accumulate the value at the current index
-        result[resultIndex] += inputData[i];
-    }
+    // Compute sum
+    const [type, result, resultDims] = reduce_helper((a, b) => a + b, input, dim, keepdim);
 
+    // Divide by number of elements in the dimension
     if (inputDims[dim] !== 1) {
         for (let i = 0; i < result.length; ++i) {
-            result[i] = result[i] / inputDims[dim];
+            result[i] /= inputDims[dim];
         }
     }
 
-    if (!keepdim) {
-        resultDims.splice(dim, 1);
-    }
-
-    return new Tensor(input.type, result, resultDims);
+    return new Tensor(type, result, resultDims);
 }
 
 
@@ -34893,6 +35348,10 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   GraniteModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GraniteModel),
 /* harmony export */   GranitePreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GranitePreTrainedModel),
 /* harmony export */   Grok1Tokenizer: () => (/* reexport safe */ _tokenizers_js__WEBPACK_IMPORTED_MODULE_3__.Grok1Tokenizer),
+/* harmony export */   GroundingDinoForObjectDetection: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GroundingDinoForObjectDetection),
+/* harmony export */   GroundingDinoImageProcessor: () => (/* reexport safe */ _models_image_processors_js__WEBPACK_IMPORTED_MODULE_13__.GroundingDinoImageProcessor),
+/* harmony export */   GroundingDinoPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GroundingDinoPreTrainedModel),
+/* harmony export */   GroundingDinoProcessor: () => (/* reexport safe */ _models_processors_js__WEBPACK_IMPORTED_MODULE_16__.GroundingDinoProcessor),
 /* harmony export */   GroupViTModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GroupViTModel),
 /* harmony export */   GroupViTPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.GroupViTPreTrainedModel),
 /* harmony export */   HerbertTokenizer: () => (/* reexport safe */ _tokenizers_js__WEBPACK_IMPORTED_MODULE_3__.HerbertTokenizer),
@@ -35122,6 +35581,7 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   RTDetrModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.RTDetrModel),
 /* harmony export */   RTDetrObjectDetectionOutput: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.RTDetrObjectDetectionOutput),
 /* harmony export */   RTDetrPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.RTDetrPreTrainedModel),
+/* harmony export */   RawAudio: () => (/* reexport safe */ _utils_audio_js__WEBPACK_IMPORTED_MODULE_5__.RawAudio),
 /* harmony export */   RawImage: () => (/* reexport safe */ _utils_image_js__WEBPACK_IMPORTED_MODULE_6__.RawImage),
 /* harmony export */   RepetitionPenaltyLogitsProcessor: () => (/* reexport safe */ _generation_logits_process_js__WEBPACK_IMPORTED_MODULE_20__.RepetitionPenaltyLogitsProcessor),
 /* harmony export */   ResNetForImageClassification: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.ResNetForImageClassification),
@@ -35187,6 +35647,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   Starcoder2PreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Starcoder2PreTrainedModel),
 /* harmony export */   StoppingCriteria: () => (/* reexport safe */ _generation_stopping_criteria_js__WEBPACK_IMPORTED_MODULE_19__.StoppingCriteria),
 /* harmony export */   StoppingCriteriaList: () => (/* reexport safe */ _generation_stopping_criteria_js__WEBPACK_IMPORTED_MODULE_19__.StoppingCriteriaList),
+/* harmony export */   StyleTextToSpeech2Model: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.StyleTextToSpeech2Model),
+/* harmony export */   StyleTextToSpeech2PreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.StyleTextToSpeech2PreTrainedModel),
 /* harmony export */   SummarizationPipeline: () => (/* reexport safe */ _pipelines_js__WEBPACK_IMPORTED_MODULE_1__.SummarizationPipeline),
 /* harmony export */   SuppressTokensAtBeginLogitsProcessor: () => (/* reexport safe */ _generation_logits_process_js__WEBPACK_IMPORTED_MODULE_20__.SuppressTokensAtBeginLogitsProcessor),
 /* harmony export */   Swin2SRForImageSuperResolution: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Swin2SRForImageSuperResolution),