Update to V0.2

app.py

@@ -1,15 +1,8 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 16:59
-# @Author   : q275343119
-# @File     : app.py
-# @Description: Main entry point for the application
 from app.ui.home import init_home
 from app.ui.about import init_about
 
 import gradio as gr
 
-
-
 home = init_home()
 about = init_about()
 
@@ -19,5 +12,3 @@ demo = gr.TabbedInterface(
     css="footer {display: none} .sort-button.svelte-1bvc1p0.svelte-1bvc1p0.svelte-1bvc1p0 {display: none;}"
 )
 demo.launch()
-
-

app/__init__.py

@@ -1,5 +1 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:13
-# @Author   : q275343119
-# @File     : __init__.py.py
-# @Description:
+

app/backend/__init__.py

@@ -1,5 +1 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:16
-# @Author   : q275343119
-# @File     : __init__.py.py
-# @Description: Backend module initialization
+

app/backend/constant.py

@@ -44,3 +44,44 @@ class EvaluationMetric(Enum):
     PRECISION_20 = "PRECISION@20"
     PRECISION_50 = "PRECISION@50"
     PRECISION_100 = "PRECISION@100"
+
+
+class EmbdDtype(Enum):
+    ALL = "all"
+    FLOAT_32 = "float32"
+    INT_8 = "int8"
+    BINARY = "binary"
+
+
+class EmbdDim(Enum):
+    OP1 = "<=1k"
+    OP2 = "1k-2k"
+    OP3 = "2k-5k"
+    OP4 = ">=5k"
+
+
+class Similarity(Enum):
+    ALL = "all"
+    COSINE = "cosine"
+    DOT = "dot"
+    EUCLIDEAN = "euclidean"
+
+
+LEADERBOARD_MAP = {
+    "text": [
+        "law",
+        "long-context",
+        "finance",
+        "conversational",
+        "tech",
+        "multilingual",
+        "code",
+        "healthcare"
+    ],
+    "multimodal": [
+        "text-to-photo",
+        "document-screenshot",
+        "figures-and-tables",
+        "text-to-text"
+    ]
+}

app/backend/data_engine.py

@@ -7,6 +7,27 @@ from typing import List
 import pandas as pd
 
 from app.backend.constant import ModelProvider
+from utils.cache_decorator import cache_with_custom_key
+
+COLUMNS = ['model_name', 'group_name', 'leaderboard', 'dataset_name',
+           'embd_dtype', 'embd_dim', 'num_params', 'max_tokens', 'similarity',
+           'query_instruct', 'corpus_instruct', 'ndcg_at_1', 'ndcg_at_3', 'ndcg_at_5',
+           'ndcg_at_10', 'ndcg_at_20',
+           'ndcg_at_50', 'ndcg_at_100', 'recall_at_1', 'recall_at_3',
+           'recall_at_5', 'recall_at_10', 'recall_at_20', 'recall_at_50',
+           'recall_at_100', 'precision_at_1', 'precision_at_3', 'precision_at_5',
+           'precision_at_10', 'precision_at_20', 'precision_at_50',
+           'precision_at_100']
+
+COLUMNS_TYPES = ["markdown", "str", 'str', 'str',
+                 'str', 'str', 'number', 'number', 'str',
+                 'str', 'str', 'number', 'number', 'number',
+                 'number', 'number',
+                 'number', 'number', 'number', 'number',
+                 'number', 'number', 'number', 'number',
+                 'number', 'number', 'number', 'number',
+                 'number', 'number', 'number',
+                 'number']
 
 
 class DataEngine:
@@ -14,14 +35,6 @@ class DataEngine:
     def __init__(self):
         self.df = self.init_dataframe()
 
-    @property
-    def leaderboards(self):
-        """
-        Get leaderboard data
-        """
-        with open('./mock_data/leaderboard.json', 'r') as f:
-            return json.load(f)
-
     @property
     def models(self):
         """
@@ -31,11 +44,11 @@ class DataEngine:
             return json.load(f)
 
     @property
-    def tasks(self):
+    def datasets(self):
         """
         Get tasks data
         """
-        with open('./mock_data/tasks.json', 'r') as f:
+        with open('./mock_data/datasets.json', 'r') as f:
             return json.load(f)
 
     @property
@@ -62,121 +75,69 @@ class DataEngine:
                 return True
         return organization in providers
 
-    def filter_df(self, leaderboard: str, task: str, providers: List, sort_key: str):
-
-        tasks = []
-        for lb in self.leaderboards:
-            if lb["name"] == leaderboard:
-                tasks = lb["tasks"]
-                break
-        df_list = []
-        for t in (filter(lambda x: x.upper() == task.upper(), tasks)):
-            datasets = []
-            for ta in self.tasks:
-                if ta["slug"].upper() == t.upper():
-                    datasets = ta["datasets"]
-                    break
-            for model in self.models:
-                if t in model["tasks"] and self._check_providers(model["organization"], providers):
-
-                    for dataset in datasets:
-                        results = self.results[dataset]
-                        for result in results:
-                            if result['model_name'] == model["model_name"]:
-                                d = result["results"]
-                                d["class"] = result["class"]
-                                d["organization"] = result["organization"]
-                                d["model_name"] = result["model_name"]
-                                df = pd.DataFrame([d])
-                                df = df[["class", "organization", "model_name", "ndcg_at_1", "ndcg_at_3", "ndcg_at_5",
-                                         "ndcg_at_10",
-                                         "ndcg_at_20", "ndcg_at_50", "ndcg_at_100", "recall_at_1", "recall_at_3",
-                                         "recall_at_5", "recall_at_10",
-                                         "recall_at_20", "recall_at_50", "recall_at_100", "precision_at_1",
-                                         "precision_at_3", "precision_at_5",
-                                         "precision_at_10", "precision_at_20", "precision_at_50", "precision_at_100"]]
-                                df_list.append(df)
-            if df_list:
-                return pd.concat(df_list).sort_values(by=sort_key.replace("@", '_at_').lower())
-            return pd.DataFrame(columns=["class", "organization", "model_name", "ndcg_at_1", "ndcg_at_3", "ndcg_at_5",
-                                         "ndcg_at_10",
-                                         "ndcg_at_20", "ndcg_at_50", "ndcg_at_100", "recall_at_1", "recall_at_3",
-                                         "recall_at_5", "recall_at_10",
-                                         "recall_at_20", "recall_at_50", "recall_at_100", "precision_at_1",
-                                         "precision_at_3", "precision_at_5",
-                                         "precision_at_10", "precision_at_20", "precision_at_50", "precision_at_100"])
-
-    def get_model_result(self, model: dict, task_datasets_map: dict, results: dict):
-        """
-        get_model_result
-        """
-        model_class = model["class"]
-        model_organization = model["organization"]
-        model_model_name = model["model_name"]
-        for leaderboard in model["leaderboards"]:
-            for task in model["tasks"]:
-                for dateset in task_datasets_map.get(task, []):
-                    for result in results[dateset]:
-                        if result["model_name"] == model_model_name:
-                            d_result = result["results"]
-                            d_result["class"] = model_class
-                            d_result["organization"] = model_organization
-                            d_result["model_name"] = model_model_name
-                            d_result["leaderboard"] = leaderboard
-                            d_result["dateset"] = dateset
-                            d_result["task"] = task
-                            yield d_result
-
+    @cache_with_custom_key("json_result")
     def jsons_to_df(self):
 
-        # change leaderboards to task_leaderboard_map
-        task_leaderboard_map = {}
-        leaderboards = self.leaderboards
-        for leaderboard in leaderboards:
-            for task in leaderboard["tasks"]:
-                task_leaderboard_map[task] = leaderboard["name"]
-
-        # change tasks to task_datasets_map
-        task_datasets_map = {}
-        for task in self.tasks:
-            task_datasets_map[task["slug"]] = task["datasets"]
-
+        results_list = self.results
         df_results_list = []
-        results = self.results
-        for model in self.models:
-            for d_result in self.get_model_result(model, task_datasets_map, results):
-                if d_result:
-                    df_results_list.append(pd.DataFrame([d_result]))
-
-        if df_results_list:
-            df_result = pd.concat(df_results_list)
-            return df_result[
-                ["leaderboard", "task", "class", "organization", "model_name", "dateset", "ndcg_at_1", "ndcg_at_3",
-                 "ndcg_at_5",
-                 "ndcg_at_10",
-                 "ndcg_at_20", "ndcg_at_50", "ndcg_at_100", "recall_at_1", "recall_at_3",
-                 "recall_at_5", "recall_at_10",
-                 "recall_at_20", "recall_at_50", "recall_at_100", "precision_at_1",
-                 "precision_at_3", "precision_at_5",
-                 "precision_at_10", "precision_at_20", "precision_at_50", "precision_at_100"]], leaderboards
-        return pd.DataFrame(
-            columns=["leaderboard", "task", "class", "organization", "model_name", "dateset", "ndcg_at_1", "ndcg_at_3",
-                     "ndcg_at_5",
-                     "ndcg_at_10",
-                     "ndcg_at_20", "ndcg_at_50", "ndcg_at_100", "recall_at_1", "recall_at_3",
-                     "recall_at_5", "recall_at_10",
-                     "recall_at_20", "recall_at_50", "recall_at_100", "precision_at_1",
-                     "precision_at_3", "precision_at_5",
-                     "precision_at_10", "precision_at_20", "precision_at_50", "precision_at_100"]), leaderboards
-
-    def filter_by_providers(self, df_result: pd.DataFrame, providers: List):
+        for result_dict in results_list:
+            dataset_name = result_dict["dataset_name"]
+            df_result_row = pd.DataFrame(result_dict["results"])
+            df_result_row["dataset_name"] = dataset_name
+            df_results_list.append(df_result_row)
+        df_result = pd.concat(df_results_list)
+
+        df_datasets_list = []
+        for item in self.datasets:
+            dataset_names = item["dataset_names"]
+            df_dataset_row = pd.DataFrame(
+                {
+                    "group_name": [item["group_name"] for _ in range(len(dataset_names))],
+                    "dataset_name": dataset_names,
+                    "leaderboard": [item["leaderboard"] for _ in range(len(dataset_names))]
+                }
+            )
+            df_datasets_list.append(df_dataset_row)
+        df_dataset = pd.concat(df_datasets_list).drop_duplicates()
+
+        models_list = self.models
+
+        df_model = pd.DataFrame(models_list)
+
+        df = pd.merge(df_result, df_model, on="model_name", how="inner")
+        df = pd.merge(df, df_dataset, on="dataset_name", how="inner")
+
+        df["model_name"] = df.apply(lambda
+                                        x: f"""<a target=\"_blank\" style=\"text-decoration: underline\" href=\"{x["reference"]}\">{x["model_name"]}</a>""",
+                                    axis=1)
+        return df[COLUMNS]
+
+    def filter_df(self, df_result: pd.DataFrame, embd_dtype: str, embd_dims: List, similarity: str, max_tokens: int):
         """
         filter_by_providers
         """
-        if not providers:
-            # providers are empty, return empty
+        if not embd_dims:
             return df_result[0:0]
-        return df_result[df_result['organization'].apply(lambda x: self._check_providers(x, providers))]
+
+        if embd_dtype and embd_dtype != "all":
+            df_result = df_result[df_result['embd_dtype'] == embd_dtype][:]
+
+        if similarity and similarity != "all":
+            df_result = df_result[df_result['similarity'] == similarity][:]
+
+        if max_tokens:
+            df_result = df_result[df_result['max_tokens'] >= max_tokens][:]
+
+        if embd_dims:
+            bins = [0, 1000, 2000, 5000, float('inf')]
+            labels = ['<=1k', '1k-2k', '2k-5k', '>=5k']
+
+            # 使用 pd.cut 进行分组
+            df_result['value_group'] = pd.cut(df_result['embd_dim'], bins=bins, labels=labels, right=False)
+            df_result = df_result[df_result['value_group'].isin(embd_dims)]
+            df_result = df_result[COLUMNS]
+
+        return df_result
 
     def summarize_dataframe(self):
         """

app/ui/__init__.py

@@ -1,5 +0,0 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:16
-# @Author   : q275343119
-# @File     : __init__.py.py
-# @Description:

app/ui/about.py

@@ -1,9 +1,3 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:20
-# @Author   : q275343119
-# @File     : about.py
-# @Description: About page implementation
-
 import gradio as gr
 
 

app/ui/component/__init__.py

@@ -1,5 +1 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:25
-# @Author   : q275343119
-# @File     : __init__.py.py
-# @Description:
+

app/ui/component/filter_component.py

@@ -1,8 +1,3 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:26
-# @Author   : q275343119
-# @File     : filter_component.py
-# @Description: Filter component implementation
 """
 Filter component for data filtering
 """
@@ -15,12 +10,27 @@ from app.backend.data_engine import DataEngine
 class FilterComponent:
 
     def __init__(self, data_engine: DataEngine, navigations: List, model_provides: List,
-                 evaluation_metrics: List):
+                 evaluation_metrics: List, embd_dtypes: List, embd_dims: List, similarities: List):
         self.data_engine = data_engine
         self.navigations = navigations
         self.task_categories = []
         self.model_provides = model_provides
         self.evaluation_metrics = evaluation_metrics
+        self.embd_dtypes = embd_dtypes
+        self.embd_dims = embd_dims
+        self.similarities = similarities
+
+    def _validate_input(self,value:str):
+        if value == "":
+            return value
+        try:
+            num = int(value)
+            if num > 0:
+                return num
+            else:
+                return ""
+        except ValueError:
+            return ""
 
     def show(self):
         with gr.Row():
@@ -40,7 +50,42 @@ class FilterComponent:
                 interactive=True,
                 elem_classes=["filter-checkbox-group"],
                 scale=2,
+                visible=False
+            )
+            embd_dtypes = gr.Dropdown(
+                label="Embd Dtype",
+                choices=self.embd_dtypes,
+                value=self.embd_dtypes[0],
+                interactive=True,
+                elem_classes=["filter-checkbox-group"],
+                scale=2,
+
+            )
+
+            embd_dims = gr.CheckboxGroup(
+                label="Embd Dim",
+                choices=self.embd_dims,
+                value=self.embd_dims,
+                interactive=True,
+                elem_classes=["filter-checkbox-group"],
+                scale=2,
             )
+
+            similarities = gr.Dropdown(
+                label="Similarity",
+                choices=self.similarities,
+                value=self.similarities[0],
+                interactive=True,
+                elem_classes=["filter-checkbox-group"],
+                scale=2,
+
+            )
+
+            max_tokens = gr.Number(
+                label="Max Tokens",
+
+            )
+
             evaluation_metrics = gr.Dropdown(
                 label="Evaluation Metrics",
                 choices=self.evaluation_metrics,
@@ -50,4 +95,5 @@ class FilterComponent:
                 scale=2,
             )
 
-        return navigations, model_provides, evaluation_metrics
+
+        return navigations, model_provides, evaluation_metrics, embd_dtypes, embd_dims, similarities, max_tokens

app/ui/component/subtabs_component.py

@@ -2,7 +2,8 @@ from typing import List
 import gradio as gr
 import pandas as pd
 
-from app.backend.data_engine import DataEngine
+from app.backend.constant import LEADERBOARD_MAP
+from app.backend.data_engine import DataEngine, COLUMNS_TYPES
 
 
 class SubtabsComponent:
@@ -10,37 +11,45 @@ class SubtabsComponent:
     def __init__(self, data_engine: DataEngine):
         self.data_engine = data_engine
 
-    def show(self, navigations: str = None, model_provides: List = None, evaluation_metrics: str = None):
-        df_result, leaderboards = self.data_engine.jsons_to_df()
+    def show(self, navigations: str = None, model_provides: List = None, evaluation_metrics: str = None,
+             embd_dtypes: str = None, embd_dims: List = None, similarity: str = None, max_tokens: str = None):
+
+        df_result = self.data_engine.jsons_to_df()
 
         navigations = "Text" if navigations is None else navigations.split(" ", maxsplit=1)[0]
-        model_provides = [] if model_provides is None else model_provides
+
         evaluation_metrics = "NDCG@10" if evaluation_metrics is None else evaluation_metrics
+        embd_dtype = "all" if embd_dtypes is None else embd_dtypes
+        embd_dims = [] if embd_dims is None else embd_dims
+        similarity = "all" if similarity is None else similarity
+        max_tokens = 0 if max_tokens is None else max_tokens
 
-        df_result = self.data_engine.filter_by_providers(df_result, model_provides)
+        df_result = self.data_engine.filter_df(df_result, embd_dtype, embd_dims, similarity, max_tokens)
 
         sort_col = evaluation_metrics.replace("@", '_at_').lower()
-        
+
         df_result = df_result.sort_values(by=sort_col, ascending=False)
 
         items = []
-        for leaderboard in leaderboards:
-            with gr.Column(visible=leaderboard["name"] == navigations) as column:
+        for leaderboard, group_names in LEADERBOARD_MAP.items():
+            with gr.Column(visible=leaderboard.upper() == navigations.upper()) as column:
                 with gr.Tabs():
                     with gr.TabItem("overall"):
-                        df_leaderboard = df_result[df_result["leaderboard"] == leaderboard["slug"]]
+                        df_leaderboard = df_result[df_result["leaderboard"] == leaderboard]
                         gr_df = gr.Dataframe(
                             df_leaderboard,
+                            datatype=COLUMNS_TYPES,
                             interactive=False
                         )
 
                         items.append(gr_df)
 
-                    for task in leaderboard["tasks"]:
-                        with gr.TabItem(task):
-                            df = df_leaderboard[df_leaderboard["task"] == task]
+                    for group_name in group_names:
+                        with gr.TabItem(group_name):
+                            df = df_leaderboard[df_leaderboard["group_name"] == group_name]
                             gr_df = gr.Dataframe(
                                 df,
+                                datatype=COLUMNS_TYPES,
                                 interactive=False
                             )
                             items.append(gr_df)

app/ui/home.py

@@ -1,13 +1,7 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:20
-# @Author   : q275343119
-# @File     : home.py
-# @Description: Home page implementation and UI components
 import pandas as pd
 
-from app.backend.constant import Navigation, TaskCategory, ModelProvider, EvaluationMetric
+from app.backend.constant import Navigation, ModelProvider, EvaluationMetric, EmbdDtype, EmbdDim, Similarity
 from app.backend.data_engine import DataEngine
-from app.ui.component.df_component import DataFrameComponent
 from app.ui.component.filter_component import FilterComponent
 from app.ui.component.subtabs_component import SubtabsComponent
 from app.ui.static import HOME_CSS
@@ -35,9 +29,12 @@ def init_home():
             [element.value for element in Navigation],
             [element.value for element in ModelProvider],
             [element.value for element in EvaluationMetric],
+            [element.value for element in EmbdDtype],
+            [element.value for element in EmbdDim],
+            [element.value for element in Similarity],
         )
 
-        navigations, model_provides, evaluation_metrics = filter_area.show()
+        navigations, model_provides, evaluation_metrics, embd_dtypes, embd_dims, similarities, max_tokens = filter_area.show()
 
         sub_tabs = SubtabsComponent(data_engine)
         columns = sub_tabs.show()
@@ -45,39 +42,24 @@ def init_home():
         # df_area = DataFrameComponent(data_engine)
         # df_display = df_area.show(pd.DataFrame(columns=[element.value for element in Navigation]))
 
-
-        gr.Markdown(f"""
-        - **Total Datasets**: {NUM_DATASETS}
-        - **Total Languages**: 113
-        - **Total Scores**: {NUM_SCORES}
-        - **Total Models**: {NUM_MODELS}
-        """ + r"""
-        Made with ❤️ for NLP. If this work is useful to you, please consider citing:
-
-        ```bibtex
-        @article{muennighoff2022mteb,
-            doi = {10.48550/ARXIV.2210.07316},
-            url = {https://arxiv.org/abs/2210.07316},
-            author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Lo{\"\i}c and Reimers, Nils},
-            title = {MTEB: Massive Text Embedding Benchmark},
-            publisher = {arXiv},
-            journal={arXiv preprint arXiv:2210.07316},  
-            year = {2022}
-        }
-        ```
-        """)
-        # all_outputs = list[sub_tab_map.values()]
         block.load(sub_tabs.show,
-                   inputs=[navigations, model_provides, evaluation_metrics], outputs=columns)
+                   inputs=[navigations, model_provides, evaluation_metrics, embd_dtypes, embd_dims, similarities,
+                           max_tokens], outputs=columns)
 
         gr.on(triggers=[navigations.change,
                         model_provides.change,
-                        evaluation_metrics.change],
-              inputs=[navigations, model_provides, evaluation_metrics],
+                        evaluation_metrics.change,
+                        embd_dtypes.change,
+                        embd_dims.change,
+                        similarities.change,
+                        max_tokens.change
+                        ],
+              inputs=[navigations, model_provides, evaluation_metrics, embd_dtypes, embd_dims, similarities,
+                      max_tokens],
               fn=sub_tabs.show,
-              outputs=columns
+              outputs=columns,
+              trigger_mode="once"
               )
 
-
     block.queue(max_size=10)
     return block

app/ui/static.py

@@ -1,9 +1,3 @@
-# -*- coding: utf-8 -*-
-# @Date     : 2024/12/24 17:16
-# @Author   : q275343119
-# @File     : static.py
-# @Description:
-
 HOME_CSS = """
 table > thead {
     white-space: normal

mock_data/datasets.json

@@ -0,0 +1,17 @@
+[
+    {"group_name": "text", "dataset_names": ["LegalQuAD", "MBPP", "PyTorch1024", "TAT-QA"], "leaderboard": "text"},
+    {"group_name": "law", "dataset_names": ["LegalQuAD"], "leaderboard": "text"},
+    {"group_name": "multilingual", "dataset_names": ["LegalQuAD"], "leaderboard": "text"},
+    {"group_name": "german", "dataset_names": ["LegalQuAD"], "leaderboard": "text"},
+    {"group_name": "code", "dataset_names": ["MBPP"], "leaderboard": "text"},
+    {"group_name": "tech", "dataset_names": ["PyTorch1024", "TAT-QA"], "leaderboard": "text"},
+    {"group_name": "long-context", "dataset_names": ["LongContextData1", "LongContextData2"], "leaderboard": "text"},
+    {"group_name": "finance", "dataset_names": ["FinanceData1", "FinanceData2"], "leaderboard": "text"},
+    {"group_name": "conversational", "dataset_names": ["ConversationalData1", "ConversationalData2"], "leaderboard": "text"},
+    {"group_name": "healthcare", "dataset_names": ["HealthcareData1", "HealthcareData2"], "leaderboard": "text"},
+    {"group_name": "multimodal", "dataset_names": ["MultimodalData1", "MultimodalData2"], "leaderboard": "multimodal"},
+    {"group_name": "text-to-photo", "dataset_names": ["TextToPhotoData1", "TextToPhotoData2"], "leaderboard": "multimodal"},
+    {"group_name": "document-screenshot", "dataset_names": ["DocumentScreenshotData1", "DocumentScreenshotData2"], "leaderboard": "multimodal"},
+    {"group_name": "figures-and-tables", "dataset_names": ["FiguresAndTablesData1", "FiguresAndTablesData2"], "leaderboard": "multimodal"},
+    {"group_name": "text-to-text", "dataset_names": ["TextToTextData1", "TextToTextData2"], "leaderboard": "multimodal"}
+]

mock_data/leaderboard.json

@@ -1,26 +0,0 @@
-[
-  {
-    "slug": "text",
-    "name": "Text",
-    "tasks": [
-      "law",
-      "long-context",
-      "finance",
-      "conversational",
-      "tech",
-      "multilingual",
-      "code",
-      "healthcare"
-    ]
-  },
-  {
-    "slug": "multimodal",
-    "name": "Multimodal",
-    "tasks": [
-      "text-to-photo",
-      "document-screenshot",
-      "figures-and-tables",
-      "text-to-text"
-    ]
-  }
-]

mock_data/models.json

@@ -1,400 +1,113 @@
 [
+    {"model_name":"sentence-transformers/all-MiniLM-L6-v2","embd_dtype":"float32","embd_dim":384,"num_params":22700000,"max_tokens":256,
+    "similarity":"cosine","query_instruct":null,"corpus_instruct":null,
+    "reference":"https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2"},
+    {"model_name":"jinaai/jina-embeddings-v2-small-en","embd_dtype":"float32","embd_dim":512,"num_params":32700000,"max_tokens":8192,
+    "similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://huggingface.co/jinaai/jina-embeddings-v2-small-en"},
 
-  {
-    "class": "sentence-transformers",
-    "organization": "BAAI",
-    "model_name": "bge-law-transformer",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "law"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "OpenAI",
-    "model_name": "openai-law-embed-v2",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "law"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Google",
-    "model_name": "google-law-model-v1",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "law"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "HuggingFace",
-    "model_name": "hf-long-context-transformer",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "long-context"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Apple",
-    "model_name": "apple-long-context-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "long-context"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Microsoft",
-    "model_name": "microsoft-long-context-v1",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "long-context"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "HuggingFace",
-    "model_name": "hf-finance-transformer",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "finance"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "OpenAI",
-    "model_name": "openai-finance-embed-v1",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "finance"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Google",
-    "model_name": "google-finance-embed-v1",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "finance"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-conversational-v2",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "conversational"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Facebook",
-    "model_name": "fb-conversational-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "conversational"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Amazon",
-    "model_name": "amazon-conversational-transformer",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "conversational"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Tesla",
-    "model_name": "tesla-tech-model",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "tech"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Snowflake",
-    "model_name": "snowflake-tech-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "tech"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "DeepMind",
-    "model_name": "dm-tech-v1",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "tech"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Tesla",
-    "model_name": "tesla-multilingual",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "multilingual"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Apple",
-    "model_name": "apple-multilingual-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "multilingual"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Google",
-    "model_name": "google-multilingual-model",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "multilingual"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-code-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "code"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "HuggingFace",
-    "model_name": "hf-code-transformer",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "code"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Apple",
-    "model_name": "apple-code-model",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "code"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-healthcare-v2",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "healthcare"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Facebook",
-    "model_name": "fb-healthcare-embed",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "healthcare"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Google",
-    "model_name": "google-healthcare-model",
-    "leaderboards": [
-      "text"
-    ],
-    "tasks": [
-      "healthcare"
-    ]
-  },
-
-  {
-    "class": "sentence-transformers",
-    "organization": "Facebook",
-    "model_name": "fb-text-to-photo",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-photo"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Apple",
-    "model_name": "apple-text-to-photo-v2",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-photo"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-text-to-photo",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-photo"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-document-screenshot",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "document-screenshot"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Google",
-    "model_name": "google-document-screenshot",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "document-screenshot"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Facebook",
-    "model_name": "fb-document-screenshot",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "document-screenshot"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "Google",
-    "model_name": "google-figures-tables",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "figures-and-tables"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Microsoft",
-    "model_name": "microsoft-figures-tables",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "figures-and-tables"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Apple",
-    "model_name": "apple-figures-tables",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "figures-and-tables"
-    ]
-  },
-  {
-    "class": "sentence-transformers",
-    "organization": "OpenAI",
-    "model_name": "openai-text-to-text",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-text"
-    ]
-  },
-  {
-    "class": "proprietary",
-    "organization": "Facebook",
-    "model_name": "fb-text-to-text",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-text"
-    ]
-  },
-  {
-    "class": "custom",
-    "organization": "Apple",
-    "model_name": "apple-text-to-text",
-    "leaderboards": [
-      "multimodal"
-    ],
-    "tasks": [
-      "text-to-text"
-    ]
-  }
-]
+    {"model_name":"text-embedding-3-small","embd_dtype":"float32","embd_dim":1536,"num_params":null,"max_tokens":8191,
+    "similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://docs.voyageai.com/docs/embeddings"},
+    {"model_name":"voyage-3-lite","embd_dtype":"float32","embd_dim":512,"num_params":null,"max_tokens":32000,"similarity":"cosine",
+    "query_instruct":"Represent the query for retrieving supporting documents: ","corpus_instruct":"Represent the document for retrieval: ",
+    "reference":"https://docs.voyageai.com/docs/embeddings"},
+    {"model_name":"mock-model-1","embd_dtype":"float32","embd_dim":512,"num_params":5000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-1"},
+    {"model_name":"mock-model-2","embd_dtype":"int8","embd_dim":256,"num_params":10000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-2"},
+    {"model_name":"mock-model-3","embd_dtype":"binary","embd_dim":1024,"num_params":2000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-3"},
+    {"model_name":"mock-model-4","embd_dtype":"float32","embd_dim":768,"num_params":3000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-4"},
+    {"model_name":"mock-model-5","embd_dtype":"int8","embd_dim":384,"num_params":7000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-5"},
+    {"model_name":"mock-model-6","embd_dtype":"binary","embd_dim":640,"num_params":15000000,"max_tokens":16384,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-6"},
+    {"model_name":"mock-model-7","embd_dtype":"float32","embd_dim":512,"num_params":2500000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-7"},
+    {"model_name":"mock-model-8","embd_dtype":"int8","embd_dim":1024,"num_params":8000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-8"},
+    {"model_name":"mock-model-9","embd_dtype":"binary","embd_dim":128,"num_params":9000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-9"},
+    {"model_name":"mock-model-10","embd_dtype":"float32","embd_dim":1536,"num_params":6000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-10"},
+    {"model_name":"mock-model-11","embd_dtype":"int8","embd_dim":512,"num_params":11000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-11"},
+    {"model_name":"mock-model-12","embd_dtype":"binary","embd_dim":256,"num_params":12000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-12"},
+    {"model_name":"mock-model-13","embd_dtype":"float32","embd_dim":1024,"num_params":13000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-13"},
+    {"model_name":"mock-model-14","embd_dtype":"int8","embd_dim":768,"num_params":14000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-14"},
+    {"model_name":"mock-model-15","embd_dtype":"binary","embd_dim":384,"num_params":15000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-15"},
+    {"model_name":"mock-model-16","embd_dtype":"float32","embd_dim":640,"num_params":16000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-16"},
+    {"model_name":"mock-model-17","embd_dtype":"int8","embd_dim":512,"num_params":17000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-17"},
+    {"model_name":"mock-model-18","embd_dtype":"binary","embd_dim":1024,"num_params":18000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-18"},
+    {"model_name":"mock-model-19","embd_dtype":"float32","embd_dim":128,"num_params":19000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-19"},
+    {"model_name":"mock-model-20","embd_dtype":"int8","embd_dim":1536,"num_params":20000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-20"},
+    {"model_name":"mock-model-21","embd_dtype":"binary","embd_dim":512,"num_params":21000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-21"},
+    {"model_name":"mock-model-22","embd_dtype":"float32","embd_dim":256,"num_params":22000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-22"},
+    {"model_name":"mock-model-23","embd_dtype":"int8","embd_dim":1024,"num_params":23000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-23"},
+    {"model_name":"mock-model-24","embd_dtype":"binary","embd_dim":768,"num_params":24000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-24"},
+    {"model_name":"mock-model-25","embd_dtype":"float32","embd_dim":384,"num_params":25000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-25"},
+    {"model_name":"mock-model-26","embd_dtype":"int8","embd_dim":640,"num_params":26000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-26"},
+    {"model_name":"mock-model-27","embd_dtype":"binary","embd_dim":512,"num_params":27000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-27"},
+    {"model_name":"mock-model-28","embd_dtype":"float32","embd_dim":1024,"num_params":28000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-28"},
+    {"model_name":"mock-model-29","embd_dtype":"int8","embd_dim":128,"num_params":29000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-29"},
+    {"model_name":"mock-model-30","embd_dtype":"binary","embd_dim":1536,"num_params":30000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-30"},
+    {"model_name":"mock-model-31","embd_dtype":"float32","embd_dim":512,"num_params":31000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-31"},
+    {"model_name":"mock-model-32","embd_dtype":"int8","embd_dim":256,"num_params":32000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-32"},
+    {"model_name":"mock-model-33","embd_dtype":"binary","embd_dim":1024,"num_params":33000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-33"},
+    {"model_name":"mock-model-34","embd_dtype":"float32","embd_dim":768,"num_params":34000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-34"},
+    {"model_name":"mock-model-35","embd_dtype":"int8","embd_dim":384,"num_params":35000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-35"},
+    {"model_name":"mock-model-36","embd_dtype":"binary","embd_dim":640,"num_params":36000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-36"},
+    {"model_name":"mock-model-37","embd_dtype":"float32","embd_dim":512,"num_params":37000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-37"},
+    {"model_name":"mock-model-38","embd_dtype":"int8","embd_dim":1024,"num_params":38000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-38"},
+    {"model_name":"mock-model-39","embd_dtype":"binary","embd_dim":128,"num_params":39000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-39"},
+    {"model_name":"mock-model-40","embd_dtype":"float32","embd_dim":1536,"num_params":40000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-40"},
+    {"model_name":"mock-model-41","embd_dtype":"int8","embd_dim":512,"num_params":41000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-41"},
+    {"model_name":"mock-model-42","embd_dtype":"binary","embd_dim":256,"num_params":42000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-42"},
+    {"model_name":"mock-model-43","embd_dtype":"float32","embd_dim":1024,"num_params":43000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-43"},
+    {"model_name":"mock-model-44","embd_dtype":"int8","embd_dim":768,"num_params":44000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-44"},
+    {"model_name":"mock-model-45","embd_dtype":"binary","embd_dim":384,"num_params":45000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-45"},
+    {"model_name":"mock-model-46","embd_dtype":"float32","embd_dim":640,"num_params":46000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-46"},
+    {"model_name":"mock-model-47","embd_dtype":"int8","embd_dim":512,"num_params":47000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-47"},
+    {"model_name":"mock-model-48","embd_dtype":"binary","embd_dim":1024,"num_params":48000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-48"},
+    {"model_name":"mock-model-49","embd_dtype":"float32","embd_dim":128,"num_params":49000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-49"},
+    {"model_name":"mock-model-50","embd_dtype":"int8","embd_dim":1536,"num_params":50000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-50"},
+    {"model_name":"mock-model-51","embd_dtype":"binary","embd_dim":512,"num_params":51000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-51"},
+    {"model_name":"mock-model-52","embd_dtype":"float32","embd_dim":256,"num_params":52000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-52"},
+    {"model_name":"mock-model-53","embd_dtype":"int8","embd_dim":1024,"num_params":53000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-53"},
+    {"model_name":"mock-model-54","embd_dtype":"binary","embd_dim":768,"num_params":54000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-54"},
+    {"model_name":"mock-model-55","embd_dtype":"float32","embd_dim":384,"num_params":55000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-55"},
+    {"model_name":"mock-model-56","embd_dtype":"int8","embd_dim":640,"num_params":56000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-56"},
+    {"model_name":"mock-model-57","embd_dtype":"binary","embd_dim":512,"num_params":57000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-57"},
+    {"model_name":"mock-model-58","embd_dtype":"float32","embd_dim":1024,"num_params":58000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-58"},
+    {"model_name":"mock-model-59","embd_dtype":"int8","embd_dim":128,"num_params":59000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-59"},
+    {"model_name":"mock-model-60","embd_dtype":"binary","embd_dim":1536,"num_params":60000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-60"},
+    {"model_name":"mock-model-61","embd_dtype":"float32","embd_dim":512,"num_params":61000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-61"},
+    {"model_name":"mock-model-62","embd_dtype":"int8","embd_dim":256,"num_params":62000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-62"},
+    {"model_name":"mock-model-63","embd_dtype":"binary","embd_dim":1024,"num_params":63000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-63"},
+    {"model_name":"mock-model-64","embd_dtype":"float32","embd_dim":768,"num_params":64000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-64"},
+    {"model_name":"mock-model-65","embd_dtype":"int8","embd_dim":384,"num_params":65000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-65"},
+    {"model_name":"mock-model-66","embd_dtype":"binary","embd_dim":640,"num_params":66000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-66"},
+    {"model_name":"mock-model-67","embd_dtype":"float32","embd_dim":512,"num_params":67000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-67"},
+    {"model_name":"mock-model-68","embd_dtype":"int8","embd_dim":1024,"num_params":68000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-68"},
+    {"model_name":"mock-model-69","embd_dtype":"binary","embd_dim":128,"num_params":69000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-69"},
+    {"model_name":"mock-model-70","embd_dtype":"float32","embd_dim":1536,"num_params":70000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-70"},
+    {"model_name":"mock-model-71","embd_dtype":"int8","embd_dim":512,"num_params":71000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-71"},
+    {"model_name":"mock-model-72","embd_dtype":"binary","embd_dim":256,"num_params":72000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-72"},
+    {"model_name":"mock-model-73","embd_dtype":"float32","embd_dim":1024,"num_params":73000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-73"},
+    {"model_name":"mock-model-74","embd_dtype":"int8","embd_dim":768,"num_params":74000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-74"},
+    {"model_name":"mock-model-75","embd_dtype":"binary","embd_dim":384,"num_params":75000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-75"},
+    {"model_name":"mock-model-76","embd_dtype":"float32","embd_dim":640,"num_params":76000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-76"},
+    {"model_name":"mock-model-77","embd_dtype":"int8","embd_dim":512,"num_params":77000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-77"},
+    {"model_name":"mock-model-78","embd_dtype":"binary","embd_dim":1024,"num_params":78000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-78"},
+    {"model_name":"mock-model-79","embd_dtype":"float32","embd_dim":128,"num_params":79000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-79"},
+    {"model_name":"mock-model-80","embd_dtype":"int8","embd_dim":1536,"num_params":80000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-80"},
+    {"model_name":"mock-model-81","embd_dtype":"binary","embd_dim":512,"num_params":81000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-81"},
+    {"model_name":"mock-model-82","embd_dtype":"float32","embd_dim":256,"num_params":82000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-82"},
+    {"model_name":"mock-model-83","embd_dtype":"int8","embd_dim":1024,"num_params":83000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-83"},
+    {"model_name":"mock-model-84","embd_dtype":"binary","embd_dim":768,"num_params":84000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-84"},
+    {"model_name":"mock-model-85","embd_dtype":"float32","embd_dim":384,"num_params":85000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-85"},
+    {"model_name":"mock-model-86","embd_dtype":"int8","embd_dim":640,"num_params":86000000,"max_tokens":1024,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-86"},
+    {"model_name":"mock-model-87","embd_dtype":"binary","embd_dim":512,"num_params":87000000,"max_tokens":2048,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-87"},
+    {"model_name":"mock-model-88","embd_dtype":"float32","embd_dim":1024,"num_params":88000000,"max_tokens":512,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-88"},
+    {"model_name":"mock-model-89","embd_dtype":"int8","embd_dim":128,"num_params":89000000,"max_tokens":16384,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-89"},
+    {"model_name":"mock-model-90","embd_dtype":"binary","embd_dim":1536,"num_params":90000000,"max_tokens":32000,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-90"},
+    {"model_name":"mock-model-91","embd_dtype":"float32","embd_dim":512,"num_params":91000000,"max_tokens":256,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-91"},
+    {"model_name":"mock-model-92","embd_dtype":"int8","embd_dim":256,"num_params":92000000,"max_tokens":8192,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-92"},
+    {"model_name":"mock-model-93","embd_dtype":"binary","embd_dim":1024,"num_params":93000000,"max_tokens":4096,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-93"},
+    {"model_name":"mock-model-94","embd_dtype":"float32","embd_dim":768,"num_params":94000000,"max_tokens":1024,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-94"},
+    {"model_name":"mock-model-95","embd_dtype":"int8","embd_dim":384,"num_params":95000000,"max_tokens":2048,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-95"},
+    {"model_name":"mock-model-96","embd_dtype":"binary","embd_dim":640,"num_params":96000000,"max_tokens":512,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-96"},
+    {"model_name":"mock-model-97","embd_dtype":"float32","embd_dim":512,"num_params":97000000,"max_tokens":16384,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-97"},
+    {"model_name":"mock-model-98","embd_dtype":"int8","embd_dim":1024,"num_params":98000000,"max_tokens":32000,"similarity":"dot","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-98"},
+    {"model_name":"mock-model-99","embd_dtype":"binary","embd_dim":128,"num_params":99000000,"max_tokens":8191,"similarity":"euclidean","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-99"},
+    {"model_name":"mock-model-100","embd_dtype":"float32","embd_dim":1536,"num_params":100000000,"max_tokens":4096,"similarity":"cosine","query_instruct":null,"corpus_instruct":null,"reference":"https://example.com/mock-model-100"}
+]

mock_data/tasks.json

@@ -1,110 +0,0 @@
-[
-  {
-    "slug": "law",
-    "name": "Law",
-    "datasets": [
-      "ALIA_casedocs",
-      "LegalTexts",
-      "CaseLaw"
-    ]
-  },
-  {
-    "slug": "finance",
-    "name": "Finance",
-    "datasets": [
-      "FinancialStatements",
-      "StockMarketData",
-      "BankingRecords"
-    ]
-  },
-  {
-    "slug": "healthcare",
-    "name": "Healthcare",
-    "datasets": [
-      "PatientRecords",
-      "MedicalImages",
-      "ClinicalTrials"
-    ]
-  },
-  {
-    "slug": "tech",
-    "name": "Technology",
-    "datasets": [
-      "TechBlogs",
-      "DeviceSpecifications",
-      "SoftwareDocumentation"
-    ]
-  },
-  {
-    "slug": "conversational",
-    "name": "Conversational",
-    "datasets": [
-      "ChatLogs",
-      "CustomerServiceInteractions",
-      "ForumPosts"
-    ]
-  },
-  {
-    "slug": "code",
-    "name": "Code",
-    "datasets": [
-      "GitHubRepositories",
-      "CodeSnippets",
-      "ProgrammingQuestions"
-    ]
-  },
-  {
-    "slug": "multilingual",
-    "name": "Multilingual",
-    "datasets": [
-      "ParallelCorpora",
-      "TranslationDatasets",
-      "MultilingualText"
-    ]
-  },
-  {
-    "slug": "long-context",
-    "name": "Long Context",
-    "datasets": [
-      "ResearchPapers",
-      "Books",
-      "LegalDocuments"
-    ]
-  },
-  {
-    "slug": "text-to-photo",
-    "name": "Text to Photo",
-    "datasets": [
-      "TextPhotoPairs",
-      "ImageCaptions",
-      "VisualDescriptions"
-    ]
-  },
-  {
-    "slug": "document-screenshot",
-    "name": "Document Screenshot",
-    "datasets": [
-      "ScannedDocuments",
-      "PDFScreenshots",
-      "DigitalForms"
-    ]
-  },
-  {
-    "slug": "figures-and-tables",
-    "name": "Figures and Tables",
-    "datasets": [
-      "ScientificFigures",
-      "DataTables",
-      "ChartsAndGraphs"
-    ]
-  },
-  {
-    "slug": "text-to-text",
-    "name": "Text to Text",
-    "datasets": [
-      "ParaphraseDatasets",
-      "SummarizationDatasets",
-      "TextRewriting"
-    ]
-  }
-]

requirements.txt

@@ -1,3 +1,4 @@
-gradio==4.20.0
+gradio==4.44.1
 uvicorn>=0.14.0
-fastapi<0.113.0
+fastapi<0.113.0
+cachetools==5.5.0

utils/cache_decorator.py

@@ -0,0 +1,25 @@
+# -*- coding: utf-8 -*-
+# @Date     : 2025/1/10 10:06
+# @Author   : q275343119
+# @File     : cache_decorator.py
+# @Description:
+from cachetools import TTLCache
+from functools import wraps
+
+CACHE = TTLCache(maxsize=5, ttl=3600)
+
+
+def cache_with_custom_key(cache_key: str):
+    def decorator(func):
+        @wraps(func)
+        def wrapper(*args, **kwargs):
+            if cache_key in CACHE:
+                return CACHE[cache_key]
+
+            result = func(*args, **kwargs)
+            CACHE[cache_key] = result
+            return result
+
+        return wrapper
+
+    return decorator