@huggingface/transformers 3.1.1 → 3.1.2

package/dist/transformers.cjs

@@ -56,10 +56,10 @@ module.exports = require("url");
 
 /***/ }),
 
-/***/ "?cb4d":
-/*!*************************************!*\
-  !*** #onnxruntime-webgpu (ignored) ***!
-  \*************************************/
+/***/ "?8b6b":
+/*!*********************************!*\
+  !*** onnxruntime-web (ignored) ***!
+  \*********************************/
 /***/ (() => {
 
 /* (ignored) */
@@ -3896,7 +3896,7 @@ const version = '1.20.1';
 
 "use strict";
 var onnxruntime_node__WEBPACK_IMPORTED_MODULE_1___namespace_cache;
-var _onnxruntime_webgpu__WEBPACK_IMPORTED_MODULE_2___namespace_cache;
+var onnxruntime_web__WEBPACK_IMPORTED_MODULE_2___namespace_cache;
 __webpack_require__.r(__webpack_exports__);
 /* harmony export */ __webpack_require__.d(__webpack_exports__, {
 /* harmony export */   Tensor: () => (/* reexport safe */ onnxruntime_common__WEBPACK_IMPORTED_MODULE_3__.Tensor),
@@ -3907,7 +3907,7 @@ __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 onnxruntime_node__WEBPACK_IMPORTED_MODULE_1__ = __webpack_require__(/*! onnxruntime-node */ "onnxruntime-node");
-/* harmony import */ var _onnxruntime_webgpu__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! #onnxruntime-webgpu */ "?cb4d");
+/* harmony import */ var onnxruntime_web__WEBPACK_IMPORTED_MODULE_2__ = __webpack_require__(/*! onnxruntime-web */ "?8b6b");
 /* harmony import */ var onnxruntime_common__WEBPACK_IMPORTED_MODULE_3__ = __webpack_require__(/*! onnxruntime-common */ "./node_modules/onnxruntime-common/dist/esm/index.js");
 /**
  * @file Handler file for choosing the correct version of ONNX Runtime, based on the environment.
@@ -3933,11 +3933,6 @@ __webpack_require__.r(__webpack_exports__);
 // In either case, we select the default export if it exists, otherwise we use the named export.
 
 
-// Use subpath-imports to ensure Node.js and browser interoperability.
-// See package.json and https://nodejs.org/api/packages.html#subpath-imports
-// for more information.
-// @ts-ignore
-
 
 
 
@@ -3979,7 +3974,7 @@ if (ORT_SYMBOL in globalThis) {
 } else if (_env_js__WEBPACK_IMPORTED_MODULE_0__.apis.IS_NODE_ENV) {
     ONNX = onnxruntime_node__WEBPACK_IMPORTED_MODULE_1__ ?? /*#__PURE__*/ (onnxruntime_node__WEBPACK_IMPORTED_MODULE_1___namespace_cache || (onnxruntime_node__WEBPACK_IMPORTED_MODULE_1___namespace_cache = __webpack_require__.t(onnxruntime_node__WEBPACK_IMPORTED_MODULE_1__, 2)));
 
-    // Updated as of ONNX Runtime 1.18.0
+    // Updated as of ONNX Runtime 1.20.1
     // The following table lists the supported versions of ONNX Runtime Node.js binding provided with pre-built binaries.
     // | EPs/Platforms | Windows x64 | Windows arm64 | Linux x64         | Linux arm64 | MacOS x64 | MacOS arm64 |
     // | ------------- | ----------- | ------------- | ----------------- | ----------- | --------- | ----------- |
@@ -4002,7 +3997,7 @@ if (ORT_SYMBOL in globalThis) {
     supportedDevices.push('cpu');
     defaultDevices = ['cpu'];
 } else {
-    ONNX = /*#__PURE__*/ (_onnxruntime_webgpu__WEBPACK_IMPORTED_MODULE_2___namespace_cache || (_onnxruntime_webgpu__WEBPACK_IMPORTED_MODULE_2___namespace_cache = __webpack_require__.t(_onnxruntime_webgpu__WEBPACK_IMPORTED_MODULE_2__, 2)));
+    ONNX = /*#__PURE__*/ (onnxruntime_web__WEBPACK_IMPORTED_MODULE_2___namespace_cache || (onnxruntime_web__WEBPACK_IMPORTED_MODULE_2___namespace_cache = __webpack_require__.t(onnxruntime_web__WEBPACK_IMPORTED_MODULE_2__, 2)));
 
     if (_env_js__WEBPACK_IMPORTED_MODULE_0__.apis.IS_WEBNN_AVAILABLE) {
         // TODO: Only push supported providers (depending on available hardware)
@@ -5597,6 +5592,7 @@ function getNormalizedConfig(config) {
             break;
         case 'llama':
         case 'olmo':
+        case 'olmo2':
         case 'mobilellm':
         case 'granite':
         case 'cohere':
@@ -5926,7 +5922,7 @@ __webpack_require__.r(__webpack_exports__);
 
 
 
-const VERSION = '3.1.1';
+const VERSION = '3.1.2';
 
 // Check if various APIs are available (depends on environment)
 const IS_BROWSER_ENV = typeof window !== "undefined" && typeof window.document !== "undefined";
@@ -8068,6 +8064,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   HubertForSequenceClassification: () => (/* binding */ HubertForSequenceClassification),
 /* harmony export */   HubertModel: () => (/* binding */ HubertModel),
 /* harmony export */   HubertPreTrainedModel: () => (/* binding */ HubertPreTrainedModel),
+/* harmony export */   IJepaForImageClassification: () => (/* binding */ IJepaForImageClassification),
+/* harmony export */   IJepaModel: () => (/* binding */ IJepaModel),
+/* harmony export */   IJepaPreTrainedModel: () => (/* binding */ IJepaPreTrainedModel),
 /* harmony export */   Idefics3ForConditionalGeneration: () => (/* binding */ Idefics3ForConditionalGeneration),
 /* harmony export */   Idefics3PreTrainedModel: () => (/* binding */ Idefics3PreTrainedModel),
 /* harmony export */   ImageMattingOutput: () => (/* binding */ ImageMattingOutput),
@@ -8159,6 +8158,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   OPTForCausalLM: () => (/* binding */ OPTForCausalLM),
 /* harmony export */   OPTModel: () => (/* binding */ OPTModel),
 /* harmony export */   OPTPreTrainedModel: () => (/* binding */ OPTPreTrainedModel),
+/* harmony export */   Olmo2ForCausalLM: () => (/* binding */ Olmo2ForCausalLM),
+/* harmony export */   Olmo2Model: () => (/* binding */ Olmo2Model),
+/* harmony export */   Olmo2PreTrainedModel: () => (/* binding */ Olmo2PreTrainedModel),
 /* harmony export */   OlmoForCausalLM: () => (/* binding */ OlmoForCausalLM),
 /* harmony export */   OlmoModel: () => (/* binding */ OlmoModel),
 /* harmony export */   OlmoPreTrainedModel: () => (/* binding */ OlmoPreTrainedModel),
@@ -8171,6 +8173,8 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   Owlv2ForObjectDetection: () => (/* binding */ Owlv2ForObjectDetection),
 /* harmony export */   Owlv2Model: () => (/* binding */ Owlv2Model),
 /* harmony export */   Owlv2PreTrainedModel: () => (/* binding */ Owlv2PreTrainedModel),
+/* harmony export */   PaliGemmaForConditionalGeneration: () => (/* binding */ PaliGemmaForConditionalGeneration),
+/* harmony export */   PaliGemmaPreTrainedModel: () => (/* binding */ PaliGemmaPreTrainedModel),
 /* harmony export */   PatchTSMixerForPrediction: () => (/* binding */ PatchTSMixerForPrediction),
 /* harmony export */   PatchTSMixerModel: () => (/* binding */ PatchTSMixerModel),
 /* harmony export */   PatchTSMixerPreTrainedModel: () => (/* binding */ PatchTSMixerPreTrainedModel),
@@ -8852,7 +8856,9 @@ async function decoderForward(self, model_inputs, is_encoder_decoder = false) {
         new_model_inputs.use_cache_branch = boolTensor(!!past_key_values);
     }
     if (session.inputNames.includes('position_ids') && new_model_inputs.attention_mask && !new_model_inputs.position_ids) {
-        new_model_inputs.position_ids = createPositionIds(new_model_inputs, past_key_values);
+        // NOTE: Handle a special case for paligemma models, where positions are 1-indexed
+        const start_index = self.config.model_type === 'paligemma' ? 1 : 0;
+        new_model_inputs.position_ids = createPositionIds(new_model_inputs, past_key_values, start_index);
     }
 
     // Unpack the `past_key_values` object into model inputs
@@ -8988,14 +8994,14 @@ async function imageTextToTextForward(self, {
  * @param {Tensor} attention_mask
  * @returns {{data: BigInt64Array, dims: number[]}}
  */
-function cumsum_masked_fill(attention_mask) {
+function cumsum_masked_fill(attention_mask, start_index = 0) {
     const [bz, seq_len] = attention_mask.dims;
     const attn_mask_data = attention_mask.data;
 
     const data = new BigInt64Array(attn_mask_data.length);
     for (let i = 0; i < bz; ++i) {
         const start = i * seq_len;
-        let sum = BigInt(0);
+        let sum = BigInt(start_index);
         for (let j = 0; j < seq_len; ++j) {
             const index = start + j;
             if (attn_mask_data[index] === 0n) {
@@ -9022,10 +9028,10 @@ function cumsum_masked_fill(attention_mask) {
  *     position_ids = position_ids[:, -input_ids.shape[1] :]
  * ```
  */
-function createPositionIds(model_inputs, past_key_values = null) {
+function createPositionIds(model_inputs, past_key_values = null, start_index = 0) {
     const { input_ids, inputs_embeds, attention_mask } = model_inputs;
 
-    const { data, dims } = cumsum_masked_fill(attention_mask);
+    const { data, dims } = cumsum_masked_fill(attention_mask, start_index);
     let position_ids = new _utils_tensor_js__WEBPACK_IMPORTED_MODULE_9__.Tensor('int64', data, dims);
     if (past_key_values) {
         const offset = -(input_ids ?? inputs_embeds).dims.at(1);
@@ -11842,6 +11848,30 @@ class Florence2ForConditionalGeneration extends Florence2PreTrainedModel {
     }
 }
 
+class PaliGemmaPreTrainedModel extends PreTrainedModel {
+    forward_params = [
+        'input_ids',
+        // 'inputs_embeds',
+        'attention_mask',
+        'pixel_values',
+        'position_ids',
+        'past_key_values',
+    ];
+}
+
+class PaliGemmaForConditionalGeneration extends PaliGemmaPreTrainedModel {
+    _merge_input_ids_with_image_features(kwargs) {
+        const vision_hidden_size = kwargs.image_features.dims.at(-1);
+        const reshaped_image_hidden_states = kwargs.image_features.view(-1, vision_hidden_size);
+
+        return default_merge_input_ids_with_image_features({
+            // @ts-ignore
+            image_token_id: this.config.image_token_index,
+            ...kwargs,
+            image_features: reshaped_image_hidden_states,
+        })
+    }
+}
 
 //////////////////////////////////////////////////
 // Idefics3 Models
@@ -12380,6 +12410,13 @@ class OlmoModel extends OlmoPreTrainedModel { }
 class OlmoForCausalLM extends OlmoPreTrainedModel { }
 //////////////////////////////////////////////////
 
+//////////////////////////////////////////////////
+// OLMo2 models
+class Olmo2PreTrainedModel extends PreTrainedModel { }
+class Olmo2Model extends Olmo2PreTrainedModel { }
+class Olmo2ForCausalLM extends Olmo2PreTrainedModel { }
+//////////////////////////////////////////////////
+
 
 //////////////////////////////////////////////////
 // Granite models
@@ -12796,6 +12833,20 @@ class ViTForImageClassification extends ViTPreTrainedModel {
 //////////////////////////////////////////////////
 
 
+//////////////////////////////////////////////////
+class IJepaPreTrainedModel extends PreTrainedModel { }
+class IJepaModel extends IJepaPreTrainedModel { }
+class IJepaForImageClassification extends IJepaPreTrainedModel {
+    /**
+     * @param {any} model_inputs
+     */
+    async _call(model_inputs) {
+        return new SequenceClassifierOutput(await super._call(model_inputs));
+    }
+}
+//////////////////////////////////////////////////
+
+
 //////////////////////////////////////////////////
 class VitPosePreTrainedModel extends PreTrainedModel { }
 
@@ -15066,6 +15117,7 @@ const MODEL_MAPPING_NAMES_ENCODER_ONLY = new Map([
     ['rt_detr', ['RTDetrModel', RTDetrModel]],
     ['table-transformer', ['TableTransformerModel', TableTransformerModel]],
     ['vit', ['ViTModel', ViTModel]],
+    ['ijepa', ['IJepaModel', IJepaModel]],
     ['pvt', ['PvtModel', PvtModel]],
     ['vit_msn', ['ViTMSNModel', ViTMSNModel]],
     ['vit_mae', ['ViTMAEModel', ViTMAEModel]],
@@ -15130,6 +15182,7 @@ const MODEL_MAPPING_NAMES_DECODER_ONLY = new Map([
     ['codegen', ['CodeGenModel', CodeGenModel]],
     ['llama', ['LlamaModel', LlamaModel]],
     ['olmo', ['OlmoModel', OlmoModel]],
+    ['olmo2', ['Olmo2Model', Olmo2Model]],
     ['mobilellm', ['MobileLLMModel', MobileLLMModel]],
     ['granite', ['GraniteModel', GraniteModel]],
     ['cohere', ['CohereModel', CohereModel]],
@@ -15221,6 +15274,7 @@ const MODEL_FOR_CAUSAL_LM_MAPPING_NAMES = new Map([
     ['codegen', ['CodeGenForCausalLM', CodeGenForCausalLM]],
     ['llama', ['LlamaForCausalLM', LlamaForCausalLM]],
     ['olmo', ['OlmoForCausalLM', OlmoForCausalLM]],
+    ['olmo2', ['Olmo2ForCausalLM', Olmo2ForCausalLM]],
     ['mobilellm', ['MobileLLMForCausalLM', MobileLLMForCausalLM]],
     ['granite', ['GraniteForCausalLM', GraniteForCausalLM]],
     ['cohere', ['CohereForCausalLM', CohereForCausalLM]],
@@ -15294,6 +15348,7 @@ const MODEL_FOR_IMAGE_TEXT_TO_TEXT_MAPPING_NAMES = new Map([
     ['florence2', ['Florence2ForConditionalGeneration', Florence2ForConditionalGeneration]],
     ['qwen2-vl', ['Qwen2VLForConditionalGeneration', Qwen2VLForConditionalGeneration]],
     ['idefics3', ['Idefics3ForConditionalGeneration', Idefics3ForConditionalGeneration]],
+    ['paligemma', ['PaliGemmaForConditionalGeneration', PaliGemmaForConditionalGeneration]],
 ]);
 
 const MODEL_FOR_DOCUMENT_QUESTION_ANSWERING_MAPPING_NAMES = new Map([
@@ -15302,6 +15357,7 @@ const MODEL_FOR_DOCUMENT_QUESTION_ANSWERING_MAPPING_NAMES = new Map([
 
 const MODEL_FOR_IMAGE_CLASSIFICATION_MAPPING_NAMES = new Map([
     ['vit', ['ViTForImageClassification', ViTForImageClassification]],
+    ['ijepa', ['IJepaForImageClassification', IJepaForImageClassification]],
     ['pvt', ['PvtForImageClassification', PvtForImageClassification]],
     ['vit_msn', ['ViTMSNForImageClassification', ViTMSNForImageClassification]],
     ['fastvit', ['FastViTForImageClassification', FastViTForImageClassification]],
@@ -18180,6 +18236,106 @@ class OwlViTProcessor extends _base_processing_utils_js__WEBPACK_IMPORTED_MODULE
 }
 
 
+/***/ }),
+
+/***/ "./src/models/paligemma/processing_paligemma.js":
+/*!******************************************************!*\
+  !*** ./src/models/paligemma/processing_paligemma.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 */   PaliGemmaProcessor: () => (/* binding */ PaliGemmaProcessor)
+/* 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");
+
+
+
+
+const IMAGE_TOKEN = "<image>";
+
+function build_string_from_input(
+    prompt,
+    bos_token,
+    image_seq_len,
+    image_token,
+    num_images,
+) {
+    return `${image_token.repeat(image_seq_len * num_images)}${bos_token}${prompt}\n`
+}
+
+class PaliGemmaProcessor 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
+    static uses_processor_config = false;
+
+    /**
+     * @typedef {import('../../utils/image.js').RawImage} RawImage
+     */
+
+    // `images` is required, `text` is optional
+    async _call(/** @type {RawImage|RawImage[]} */ images, text = null, kwargs = {}) {
+        if (!text) {
+            console.warn(
+                "You are using PaliGemma without a text prefix. It will perform as a picture-captioning model."
+            )
+            text = ""
+        }
+
+        if (!Array.isArray(images)) {
+            images = [images]
+        }
+
+        if (!Array.isArray(text)) {
+            text = [text]
+        }
+
+        const bos_token = this.tokenizer.bos_token;
+        const image_seq_length = this.image_processor.config.image_seq_length;
+        let input_strings;
+        if (text.some((t) => t.includes(IMAGE_TOKEN))) {
+            input_strings = text.map(
+                sample => {
+                    const expanded_sample = sample.replaceAll(IMAGE_TOKEN, IMAGE_TOKEN.repeat(image_seq_length));
+                    const bos_rfind_index = expanded_sample.lastIndexOf(IMAGE_TOKEN);
+                    const bos_index = bos_rfind_index === -1 ? 0 : bos_rfind_index + IMAGE_TOKEN.length;
+                    return expanded_sample.slice(0, bos_index) + bos_token + expanded_sample.slice(bos_index) + "\n";
+                }
+            )
+        } else {
+            console.warn(
+                "You are passing both `text` and `images` to `PaliGemmaProcessor`. The processor expects special " +
+                "image tokens in the text, as many tokens as there are images per each text. It is recommended to " +
+                "add `<image>` tokens in the very beginning of your text. For this call, we will infer how many images " +
+                "each text has and add special tokens."
+            )
+
+            input_strings = text.map(
+                sample => build_string_from_input(
+                    sample,
+                    bos_token,
+                    image_seq_length,
+                    IMAGE_TOKEN,
+                    images.length,
+                )
+            )
+        }
+
+        const text_inputs = this.tokenizer(input_strings, kwargs);
+        const image_inputs = await this.image_processor(images, kwargs);
+
+        return {
+            ...image_inputs,
+            ...text_inputs,
+        }
+    }
+}
+
+
 /***/ }),
 
 /***/ "./src/models/processors.js":
@@ -18196,13 +18352,14 @@ __webpack_require__.r(__webpack_exports__);
 /* 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_1__.MgpstrProcessor),
 /* harmony export */   OwlViTProcessor: () => (/* reexport safe */ _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_5__.OwlViTProcessor),
-/* harmony export */   PyAnnoteProcessor: () => (/* reexport safe */ _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_6__.PyAnnoteProcessor),
-/* harmony export */   Qwen2VLProcessor: () => (/* reexport safe */ _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_7__.Qwen2VLProcessor),
-/* harmony export */   SamProcessor: () => (/* reexport safe */ _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_8__.SamProcessor),
-/* harmony export */   SpeechT5Processor: () => (/* reexport safe */ _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_9__.SpeechT5Processor),
+/* harmony export */   PaliGemmaProcessor: () => (/* reexport safe */ _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_6__.PaliGemmaProcessor),
+/* harmony export */   PyAnnoteProcessor: () => (/* reexport safe */ _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_7__.PyAnnoteProcessor),
+/* harmony export */   Qwen2VLProcessor: () => (/* reexport safe */ _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_8__.Qwen2VLProcessor),
+/* harmony export */   SamProcessor: () => (/* reexport safe */ _sam_processing_sam_js__WEBPACK_IMPORTED_MODULE_9__.SamProcessor),
+/* harmony export */   SpeechT5Processor: () => (/* reexport safe */ _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_10__.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_10__.Wav2Vec2ProcessorWithLM),
-/* harmony export */   WhisperProcessor: () => (/* reexport safe */ _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_11__.WhisperProcessor)
+/* harmony export */   Wav2Vec2ProcessorWithLM: () => (/* reexport safe */ _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_11__.Wav2Vec2ProcessorWithLM),
+/* harmony export */   WhisperProcessor: () => (/* reexport safe */ _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_12__.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");
@@ -18210,12 +18367,14 @@ __webpack_require__.r(__webpack_exports__);
 /* 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 _owlvit_processing_owlvit_js__WEBPACK_IMPORTED_MODULE_5__ = __webpack_require__(/*! ./owlvit/processing_owlvit.js */ "./src/models/owlvit/processing_owlvit.js");
-/* harmony import */ var _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_6__ = __webpack_require__(/*! ./pyannote/processing_pyannote.js */ "./src/models/pyannote/processing_pyannote.js");
-/* harmony import */ var _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_7__ = __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_8__ = __webpack_require__(/*! ./sam/processing_sam.js */ "./src/models/sam/processing_sam.js");
-/* harmony import */ var _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_9__ = __webpack_require__(/*! ./speecht5/processing_speecht5.js */ "./src/models/speecht5/processing_speecht5.js");
-/* harmony import */ var _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_10__ = __webpack_require__(/*! ./wav2vec2/processing_wav2vec2.js */ "./src/models/wav2vec2/processing_wav2vec2.js");
-/* harmony import */ var _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./whisper/processing_whisper.js */ "./src/models/whisper/processing_whisper.js");
+/* harmony import */ var _paligemma_processing_paligemma_js__WEBPACK_IMPORTED_MODULE_6__ = __webpack_require__(/*! ./paligemma/processing_paligemma.js */ "./src/models/paligemma/processing_paligemma.js");
+/* harmony import */ var _pyannote_processing_pyannote_js__WEBPACK_IMPORTED_MODULE_7__ = __webpack_require__(/*! ./pyannote/processing_pyannote.js */ "./src/models/pyannote/processing_pyannote.js");
+/* harmony import */ var _qwen2_vl_processing_qwen2_vl_js__WEBPACK_IMPORTED_MODULE_8__ = __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_9__ = __webpack_require__(/*! ./sam/processing_sam.js */ "./src/models/sam/processing_sam.js");
+/* harmony import */ var _speecht5_processing_speecht5_js__WEBPACK_IMPORTED_MODULE_10__ = __webpack_require__(/*! ./speecht5/processing_speecht5.js */ "./src/models/speecht5/processing_speecht5.js");
+/* harmony import */ var _wav2vec2_processing_wav2vec2_js__WEBPACK_IMPORTED_MODULE_11__ = __webpack_require__(/*! ./wav2vec2/processing_wav2vec2.js */ "./src/models/wav2vec2/processing_wav2vec2.js");
+/* harmony import */ var _whisper_processing_whisper_js__WEBPACK_IMPORTED_MODULE_12__ = __webpack_require__(/*! ./whisper/processing_whisper.js */ "./src/models/whisper/processing_whisper.js");
+
 
 
 
@@ -26133,6 +26292,12 @@ class PreTrainedTokenizer extends _utils_generic_js__WEBPACK_IMPORTED_MODULE_0__
         this.unk_token = this.getToken('unk_token');
         this.unk_token_id = this.model.tokens_to_ids.get(this.unk_token);
 
+        this.bos_token = this.getToken('bos_token');
+        this.bos_token_id = this.model.tokens_to_ids.get(this.bos_token);
+
+        this.eos_token = this.getToken('eos_token');
+        this.eos_token_id = this.model.tokens_to_ids.get(this.eos_token);
+
         this.model_max_length = tokenizerConfig.model_max_length;
 
         /** @type {boolean} Whether or not to strip the text when tokenizing (removing excess spaces before and after the string). */
@@ -27105,6 +27270,11 @@ class WhisperTokenizer extends PreTrainedTokenizer {
         let chunk = new_chunk();
         let time_offset = 0.0;
         const timestamp_begin = this.timestamp_begin;
+        // Whisper timestamp tokens start from 0.00 and go to timestamp 30.00 in 0.02 increments. 
+        // We can calculate the last time stamp token as timestamp_begin plus the number of tokens 
+        // tokens from 0.00 to 30.00 which is 1500. 
+        const total_timestamp_tokens = 1500; // (30.00 - 0.00) / 0.02
+        const timestamp_end = timestamp_begin + total_timestamp_tokens;
 
         let previous_tokens = [];
         let previous_token_timestamps = [];
@@ -27192,7 +27362,7 @@ class WhisperTokenizer extends PreTrainedTokenizer {
                     } else {
                         // 2/ This is a regular special token, ignoring it
                     }
-                } else if (token >= timestamp_begin) {
+                } else if (token >= timestamp_begin && token <= timestamp_end) {
                     // 3/ Timestamp token
                     const time = (token - timestamp_begin) * time_precision + time_offset;
                     const rounded_time = (0,_utils_maths_js__WEBPACK_IMPORTED_MODULE_3__.round)(time, 2);
@@ -28684,15 +28854,45 @@ __webpack_require__.r(__webpack_exports__);
  */
 
 /**
- * @typedef {Object} ProgressInfo
- * @property {'initiate' | 'download' | 'progress' | 'done'} status The status of the progress item.
- * @property {string} name This can be either:
- * - a string, the *model id* of a model repo on huggingface.co.
- * - a path to a *directory* potentially containing the file.
- * @property {string} file The name of the file
- * @property {number} [progress] A number between 0 and 100. Only available for the 'progress' status.
- * @property {number} [loaded] The number of bytes loaded. Only available for the 'progress' status.
- * @property {number} [total] The total number of bytes to be loaded. Only available for the 'progress' status.
+ * @typedef {Object} InitiateProgressInfo
+ * @property {'initiate'} status
+ * @property {string} name The model id or directory path.
+ * @property {string} file The name of the file.
+ */
+
+/**
+ * @typedef {Object} DownloadProgressInfo
+ * @property {'download'} status
+ * @property {string} name The model id or directory path.
+ * @property {string} file The name of the file.
+ */
+
+/**
+ * @typedef {Object} ProgressStatusInfo
+ * @property {'progress'} status
+ * @property {string} name The model id or directory path.
+ * @property {string} file The name of the file.
+ * @property {number} progress A number between 0 and 100.
+ * @property {number} loaded The number of bytes loaded.
+ * @property {number} total The total number of bytes to be loaded.
+ */
+
+/**
+ * @typedef {Object} DoneProgressInfo
+ * @property {'done'} status
+ * @property {string} name The model id or directory path.
+ * @property {string} file The name of the file.
+ */
+
+/**
+ * @typedef {Object} ReadyProgressInfo
+ * @property {'ready'} status
+ * @property {string} task The loaded task.
+ * @property {string} model The loaded model.
+ */
+
+/**
+ * @typedef {InitiateProgressInfo | DownloadProgressInfo | ProgressStatusInfo | DoneProgressInfo | ReadyProgressInfo} ProgressInfo
  */
 
 /**
@@ -30035,13 +30235,6 @@ async function getModelFile(path_or_repo_id, filename, fatal = true, options = {
         file: filename
     })
 
-    /** @type {import('./core.js').ProgressInfo} */
-    const progressInfo = {
-        status: 'progress',
-        name: path_or_repo_id,
-        file: filename
-    }
-
     /** @type {Uint8Array} */
     let buffer;
 
@@ -30061,7 +30254,9 @@ async function getModelFile(path_or_repo_id, filename, fatal = true, options = {
 
         // For completeness, we still fire the final progress callback
         (0,_core_js__WEBPACK_IMPORTED_MODULE_3__.dispatchCallback)(options.progress_callback, {
-            ...progressInfo,
+            status: 'progress',
+            name: path_or_repo_id,
+            file: filename,
             progress: 100,
             loaded: buffer.length,
             total: buffer.length,
@@ -30069,7 +30264,9 @@ async function getModelFile(path_or_repo_id, filename, fatal = true, options = {
     } else {
         buffer = await readResponse(response, data => {
             (0,_core_js__WEBPACK_IMPORTED_MODULE_3__.dispatchCallback)(options.progress_callback, {
-                ...progressInfo,
+                status: 'progress',
+                name: path_or_repo_id,
+                file: filename,
                 ...data,
             })
         })
@@ -30126,12 +30323,11 @@ async function getModelJSON(modelPath, fileName, fatal = true, options = {}) {
 
     return JSON.parse(jsonData);
 }
-
 /**
  * Read and track progress when reading a Response object
  *
- * @param {any} response The Response object to read
- * @param {function} progress_callback The function to call with progress updates
+ * @param {Response|FileResponse} response The Response object to read
+ * @param {(data: {progress: number, loaded: number, total: number}) => void} progress_callback The function to call with progress updates
  * @returns {Promise<Uint8Array>} A Promise that resolves with the Uint8Array buffer
  */
 async function readResponse(response, progress_callback) {
@@ -30528,6 +30724,46 @@ class RawImage {
         return this._update(newData, this.width, this.height, 4);
     }
 
+    /**
+     * Apply an alpha mask to the image. Operates in place.
+     * @param {RawImage} mask The mask to apply. It should have a single channel.
+     * @returns {RawImage} The masked image.
+     * @throws {Error} If the mask is not the same size as the image.
+     * @throws {Error} If the image does not have 4 channels.
+     * @throws {Error} If the mask is not a single channel.
+     */
+    putAlpha(mask) {
+        if (mask.width !== this.width || mask.height !== this.height) {
+            throw new Error(`Expected mask size to be ${this.width}x${this.height}, but got ${mask.width}x${mask.height}`);
+        }
+        if (mask.channels !== 1) {
+            throw new Error(`Expected mask to have 1 channel, but got ${mask.channels}`);
+        }
+
+        const this_data = this.data;
+        const mask_data = mask.data;
+        const num_pixels = this.width * this.height;
+        if (this.channels === 3) {
+            // Convert to RGBA and simultaneously apply mask to alpha channel
+            const newData = new Uint8ClampedArray(num_pixels * 4);
+            for (let i = 0, in_offset = 0, out_offset = 0; i < num_pixels; ++i) {
+                newData[out_offset++] = this_data[in_offset++];
+                newData[out_offset++] = this_data[in_offset++];
+                newData[out_offset++] = this_data[in_offset++];
+                newData[out_offset++] = mask_data[i];
+            }
+            return this._update(newData, this.width, this.height, 4);
+
+        } else if (this.channels === 4) {
+            // Apply mask to alpha channel in place
+            for (let i = 0; i < num_pixels; ++i) {
+                this_data[4 * i + 3] = mask_data[i];
+            }
+            return this;
+        }
+        throw new Error(`Expected image to have 3 or 4 channels, but got ${this.channels}`);
+    }
+
     /**
      * Resize the image to the given dimensions. This method uses the canvas API to perform the resizing.
      * @param {number} width The width of the new image. `null` or `-1` will preserve the aspect ratio.
@@ -33694,7 +33930,7 @@ function quantize_embeddings(tensor, precision) {
 /******/ 	
 /************************************************************************/
 var __webpack_exports__ = {};
-// This entry need to be wrapped in an IIFE because it need to be in strict mode.
+// This entry needs to be wrapped in an IIFE because it needs to be in strict mode.
 (() => {
 "use strict";
 /*!*****************************!*\
@@ -33963,6 +34199,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   HubertForSequenceClassification: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.HubertForSequenceClassification),
 /* harmony export */   HubertModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.HubertModel),
 /* harmony export */   HubertPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.HubertPreTrainedModel),
+/* harmony export */   IJepaForImageClassification: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.IJepaForImageClassification),
+/* harmony export */   IJepaModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.IJepaModel),
+/* harmony export */   IJepaPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.IJepaPreTrainedModel),
 /* harmony export */   Idefics3ForConditionalGeneration: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Idefics3ForConditionalGeneration),
 /* harmony export */   Idefics3ImageProcessor: () => (/* reexport safe */ _models_image_processors_js__WEBPACK_IMPORTED_MODULE_13__.Idefics3ImageProcessor),
 /* harmony export */   Idefics3PreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Idefics3PreTrainedModel),
@@ -34101,6 +34340,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   OPTModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.OPTModel),
 /* harmony export */   OPTPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.OPTPreTrainedModel),
 /* harmony export */   ObjectDetectionPipeline: () => (/* reexport safe */ _pipelines_js__WEBPACK_IMPORTED_MODULE_1__.ObjectDetectionPipeline),
+/* harmony export */   Olmo2ForCausalLM: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Olmo2ForCausalLM),
+/* harmony export */   Olmo2Model: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Olmo2Model),
+/* harmony export */   Olmo2PreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Olmo2PreTrainedModel),
 /* harmony export */   OlmoForCausalLM: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.OlmoForCausalLM),
 /* harmony export */   OlmoModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.OlmoModel),
 /* harmony export */   OlmoPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.OlmoPreTrainedModel),
@@ -34117,6 +34359,9 @@ __webpack_require__.r(__webpack_exports__);
 /* harmony export */   Owlv2ImageProcessor: () => (/* reexport safe */ _models_image_processors_js__WEBPACK_IMPORTED_MODULE_13__.Owlv2ImageProcessor),
 /* harmony export */   Owlv2Model: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Owlv2Model),
 /* harmony export */   Owlv2PreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.Owlv2PreTrainedModel),
+/* harmony export */   PaliGemmaForConditionalGeneration: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.PaliGemmaForConditionalGeneration),
+/* harmony export */   PaliGemmaPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.PaliGemmaPreTrainedModel),
+/* harmony export */   PaliGemmaProcessor: () => (/* reexport safe */ _models_processors_js__WEBPACK_IMPORTED_MODULE_16__.PaliGemmaProcessor),
 /* harmony export */   PatchTSMixerForPrediction: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.PatchTSMixerForPrediction),
 /* harmony export */   PatchTSMixerModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.PatchTSMixerModel),
 /* harmony export */   PatchTSMixerPreTrainedModel: () => (/* reexport safe */ _models_js__WEBPACK_IMPORTED_MODULE_2__.PatchTSMixerPreTrainedModel),
@@ -34452,7 +34697,7 @@ __webpack_require__.r(__webpack_exports__);
 })();
 
 var __webpack_export_target__ = exports;
-for(var i in __webpack_exports__) __webpack_export_target__[i] = __webpack_exports__[i];
+for(var __webpack_i__ in __webpack_exports__) __webpack_export_target__[__webpack_i__] = __webpack_exports__[__webpack_i__];
 if(__webpack_exports__.__esModule) Object.defineProperty(__webpack_export_target__, "__esModule", { value: true });
 /******/ })()
 ;