mteb 2.5.2__py3-none-any.whl → 2.7.9__py3-none-any.whl

mteb/models/model_implementations/dino_models.py

@@ -1,13 +1,18 @@
-from typing import Any, Literal
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any, Literal
 
 import torch
-from torch.utils.data import DataLoader
 from tqdm.auto import tqdm
 
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_meta import ModelMeta, ScoringFunction
-from mteb.types import Array, BatchedInput, PromptType
+
+if TYPE_CHECKING:
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput, PromptType
 
 
 class DINOModel(AbsEncoder):
@@ -104,7 +109,7 @@ dinov2_training_datasets = set(
 
 
 dinov2_small = ModelMeta(
-    loader=DINOModel,  # type: ignore
+    loader=DINOModel,
     name="facebook/dinov2-small",
     model_type=["dense"],
     languages=["eng-Latn"],
@@ -112,6 +117,7 @@ dinov2_small = ModelMeta(
     release_date="2023-07-18",
     modalities=["image"],
     n_parameters=22_100_000,
+    n_embedding_parameters=None,
     memory_usage_mb=84,
     max_tokens=None,
     embed_dim=384,
@@ -119,13 +125,13 @@ dinov2_small = ModelMeta(
     open_weights=True,
     public_training_code="https://github.com/facebookresearch/dinov2",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/dinov2-small",
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=False,
     training_datasets=dinov2_training_datasets,
     citation="""@misc{oquab2023dinov2,
-      title={DINOv2: Learning Robust Visual Features without Supervision}, 
+      title={DINOv2: Learning Robust Visual Features without Supervision},
       author={Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel Synnaeve and Hu Xu and Hervé Jegou and Julien Mairal and Patrick Labatut and Armand Joulin and Piotr Bojanowski},
       year={2023},
       eprint={2304.07193},
@@ -135,7 +141,7 @@ dinov2_small = ModelMeta(
 )
 
 dinov2_base = ModelMeta(
-    loader=DINOModel,  # type: ignore
+    loader=DINOModel,
     name="facebook/dinov2-base",
     model_type=["dense"],
     languages=["eng-Latn"],
@@ -143,6 +149,7 @@ dinov2_base = ModelMeta(
     release_date="2023-07-18",
     modalities=["image"],
     n_parameters=86_600_000,
+    n_embedding_parameters=None,
     memory_usage_mb=330,
     max_tokens=None,
     embed_dim=768,
@@ -150,13 +157,13 @@ dinov2_base = ModelMeta(
     open_weights=True,
     public_training_code="https://github.com/facebookresearch/dinov2",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/dinov2-base",
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=False,
     training_datasets=dinov2_training_datasets,
     citation="""@misc{oquab2023dinov2,
-      title={DINOv2: Learning Robust Visual Features without Supervision}, 
+      title={DINOv2: Learning Robust Visual Features without Supervision},
       author={Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel Synnaeve and Hu Xu and Hervé Jegou and Julien Mairal and Patrick Labatut and Armand Joulin and Piotr Bojanowski},
       year={2023},
       eprint={2304.07193},
@@ -166,7 +173,7 @@ dinov2_base = ModelMeta(
 )
 
 dinov2_large = ModelMeta(
-    loader=DINOModel,  # type: ignore
+    loader=DINOModel,
     name="facebook/dinov2-large",
     model_type=["dense"],
     languages=["eng-Latn"],
@@ -174,6 +181,7 @@ dinov2_large = ModelMeta(
     release_date="2023-07-18",
     modalities=["image"],
     n_parameters=304_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=1161,
     max_tokens=None,
     embed_dim=1024,
@@ -181,13 +189,13 @@ dinov2_large = ModelMeta(
     open_weights=True,
     public_training_code="https://github.com/facebookresearch/dinov2",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/dinov2-large",
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=False,
     training_datasets=dinov2_training_datasets,
     citation="""@misc{oquab2023dinov2,
-      title={DINOv2: Learning Robust Visual Features without Supervision}, 
+      title={DINOv2: Learning Robust Visual Features without Supervision},
       author={Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel Synnaeve and Hu Xu and Hervé Jegou and Julien Mairal and Patrick Labatut and Armand Joulin and Piotr Bojanowski},
       year={2023},
       eprint={2304.07193},
@@ -197,7 +205,7 @@ dinov2_large = ModelMeta(
 )
 
 dinov2_giant = ModelMeta(
-    loader=DINOModel,  # type: ignore
+    loader=DINOModel,
     name="facebook/dinov2-giant",
     model_type=["dense"],
     languages=["eng-Latn"],
@@ -205,6 +213,7 @@ dinov2_giant = ModelMeta(
     release_date="2023-07-18",
     modalities=["image"],
     n_parameters=1_140_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=4335,
     max_tokens=None,
     embed_dim=1536,
@@ -212,13 +221,13 @@ dinov2_giant = ModelMeta(
     open_weights=True,
     public_training_code="https://github.com/facebookresearch/dinov2",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/dinov2-giant",
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=False,
     training_datasets=dinov2_training_datasets,
     citation="""@misc{oquab2023dinov2,
-      title={DINOv2: Learning Robust Visual Features without Supervision}, 
+      title={DINOv2: Learning Robust Visual Features without Supervision},
       author={Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel Synnaeve and Hu Xu and Hervé Jegou and Julien Mairal and Patrick Labatut and Armand Joulin and Piotr Bojanowski},
       year={2023},
       eprint={2304.07193},
@@ -240,6 +249,7 @@ webssl_dino300m_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=304_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=1158,
     max_tokens=None,
     embed_dim=1024,
@@ -247,13 +257,13 @@ webssl_dino300m_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino300m-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -271,6 +281,7 @@ webssl_dino1b_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=1_130_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=4329,
     max_tokens=None,
     embed_dim=1536,
@@ -278,13 +289,13 @@ webssl_dino1b_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino1b-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -302,6 +313,7 @@ webssl_dino2b_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=2_080_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=7951,
     max_tokens=None,
     embed_dim=2688,
@@ -309,13 +321,13 @@ webssl_dino2b_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino2b-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -333,6 +345,7 @@ webssl_dino3b_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=3_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=11247,
     max_tokens=None,
     embed_dim=3072,
@@ -340,13 +353,13 @@ webssl_dino3b_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino3b-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -364,6 +377,7 @@ webssl_dino5b_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=5_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=18838,
     max_tokens=None,
     embed_dim=3584,
@@ -371,13 +385,13 @@ webssl_dino5b_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino5b-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -395,6 +409,7 @@ webssl_dino7b_full8b_224 = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=7_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=24605,
     max_tokens=None,
     embed_dim=4096,
@@ -402,13 +417,13 @@ webssl_dino7b_full8b_224 = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino7b-full8b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -426,6 +441,7 @@ webssl_dino7b_full8b_378 = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=7_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=24613,
     max_tokens=None,
     embed_dim=4096,
@@ -433,13 +449,13 @@ webssl_dino7b_full8b_378 = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino7b-full8b-378",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -457,6 +473,7 @@ webssl_dino7b_full8b_518 = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=7_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=24623,
     max_tokens=None,
     embed_dim=4096,
@@ -464,13 +481,13 @@ webssl_dino7b_full8b_518 = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino7b-full8b-518",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -489,6 +506,7 @@ webssl_dino2b_light2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=2_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=7951,
     max_tokens=None,
     embed_dim=2688,
@@ -496,13 +514,13 @@ webssl_dino2b_light2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino2b-light2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -520,6 +538,7 @@ webssl_dino2b_heavy2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=2_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=7951,
     max_tokens=None,
     embed_dim=2688,
@@ -527,13 +546,13 @@ webssl_dino2b_heavy2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino2b-heavy2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -551,6 +570,7 @@ webssl_dino3b_light2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=3_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=11247,
     max_tokens=None,
     embed_dim=3072,
@@ -558,13 +578,13 @@ webssl_dino3b_light2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino3b-light2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -582,6 +602,7 @@ webssl_dino3b_heavy2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=3_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=11247,
     max_tokens=None,
     embed_dim=3072,
@@ -589,13 +610,13 @@ webssl_dino3b_heavy2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-dino3b-heavy2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -613,6 +634,7 @@ webssl_mae300m_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=304_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=1161,
     max_tokens=None,
     embed_dim=1024,
@@ -620,13 +642,13 @@ webssl_mae300m_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-mae300m-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -644,6 +666,7 @@ webssl_mae700m_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=700_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=2412,
     max_tokens=None,
     embed_dim=1280,
@@ -651,13 +674,13 @@ webssl_mae700m_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-mae700m-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},
@@ -675,6 +698,7 @@ webssl_mae1b_full2b = ModelMeta(
     release_date="2025-04-24",
     modalities=["image"],
     n_parameters=1_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=4337,
     max_tokens=None,
     embed_dim=1536,
@@ -682,13 +706,13 @@ webssl_mae1b_full2b = ModelMeta(
     open_weights=True,
     public_training_code="",
     public_training_data=None,
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/facebook/webssl-mae1b-full2b-224",
     similarity_fn_name=None,
     use_instructions=False,
     training_datasets=webssl_dino_training_datasets,
     citation="""@article{fan2025scaling,
-  title={Scaling Language-Free Visual Representation Learning}, 
+  title={Scaling Language-Free Visual Representation Learning},
   author={David Fan and Shengbang Tong and Jiachen Zhu and Koustuv Sinha and Zhuang Liu and Xinlei Chen and Michael Rabbat and Nicolas Ballas and Yann LeCun and Amir Bar and Saining Xie},
   year={2025},
   eprint={2504.01017},

mteb/models/model_implementations/e5_instruct.py

@@ -45,11 +45,19 @@ e5_instruct = ModelMeta(
     open_weights=True,
     revision="baa7be480a7de1539afce709c8f13f833a510e0a",
     release_date=E5_PAPER_RELEASE_DATE,
-    framework=["GritLM", "PyTorch", "Sentence Transformers"],
+    framework=[
+        "GritLM",
+        "PyTorch",
+        "Sentence Transformers",
+        "ONNX",
+        "safetensors",
+        "Transformers",
+    ],
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=True,
     reference="https://huggingface.co/intfloat/multilingual-e5-large-instruct",
     n_parameters=560_000_000,
+    n_embedding_parameters=256_002_048,
     memory_usage_mb=1068,
     embed_dim=1024,
     license="mit",
@@ -84,11 +92,18 @@ e5_mistral = ModelMeta(
     open_weights=True,
     revision="07163b72af1488142a360786df853f237b1a3ca1",
     release_date=E5_PAPER_RELEASE_DATE,
-    framework=["GritLM", "PyTorch", "Sentence Transformers"],
+    framework=[
+        "GritLM",
+        "PyTorch",
+        "Sentence Transformers",
+        "safetensors",
+        "Transformers",
+    ],
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=True,
     reference="https://huggingface.co/intfloat/e5-mistral-7b-instruct",
     n_parameters=7_111_000_000,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=13563,
     embed_dim=4096,
     license="mit",
@@ -132,6 +147,7 @@ zeta_alpha_ai__zeta_alpha_e5_mistral = ModelMeta(
     release_date="2024-08-30",
     languages=["eng-Latn"],
     n_parameters=7110660096,
+    n_embedding_parameters=None,
     memory_usage_mb=13563,
     max_tokens=32768.0,
     embed_dim=4096,
@@ -139,7 +155,13 @@ zeta_alpha_ai__zeta_alpha_e5_mistral = ModelMeta(
     open_weights=True,
     public_training_data=None,
     public_training_code=None,
-    framework=["PyTorch", "Sentence Transformers", "GritLM"],
+    framework=[
+        "PyTorch",
+        "Sentence Transformers",
+        "GritLM",
+        "safetensors",
+        "Transformers",
+    ],
     reference="https://huggingface.co/zeta-alpha-ai/Zeta-Alpha-E5-Mistral",
     similarity_fn_name=ScoringFunction.COSINE,
     use_instructions=True,
@@ -209,6 +231,7 @@ BeastyZ__e5_R_mistral_7b = ModelMeta(
     release_date="2024-06-28",
     languages=["eng-Latn"],
     n_parameters=7241732096,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=27625,
     max_tokens=32768.0,
     embed_dim=4096,
@@ -216,7 +239,7 @@ BeastyZ__e5_R_mistral_7b = ModelMeta(
     open_weights=True,
     public_training_code="https://github.com/LeeSureman/E5-Retrieval-Reproduction",
     public_training_data="https://huggingface.co/datasets/BeastyZ/E5-R",
-    framework=["PyTorch"],
+    framework=["PyTorch", "Transformers", "safetensors"],
     reference="https://huggingface.co/BeastyZ/e5-R-mistral-7b",
     similarity_fn_name="cosine",
     use_instructions=True,