mteb 2.7.2__py3-none-any.whl → 2.7.9__py3-none-any.whl

mteb/models/model_implementations/rerankers_monot5_based.py

@@ -1,15 +1,21 @@
+from __future__ import annotations
+
 import logging
-from typing import Any
+from typing import TYPE_CHECKING, Any
 
 import torch
-from torch.utils.data import DataLoader
 
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.model_meta import ModelMeta
-from mteb.types import Array, BatchedInput, PromptType
 
 from .rerankers_custom import RerankerWrapper
 
+if TYPE_CHECKING:
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput, PromptType
+
+
 logger = logging.getLogger(__name__)
 
 
@@ -321,6 +327,7 @@ monot5_small = ModelMeta(
     revision="77f8e3f7b1eb1afe353aa21a7c3a2fc8feca702e",
     release_date="2022-03-28",
     n_parameters=None,
+    n_embedding_parameters=16_449_536,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -363,6 +370,7 @@ monot5_base = ModelMeta(
       url={https://arxiv.org/abs/2206.02873},
     }""",
     n_parameters=None,
+    n_embedding_parameters=24_674_304,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -387,6 +395,7 @@ monot5_large = ModelMeta(
     revision="48cfad1d8dd587670393f27ee8ec41fde63e3d98",
     release_date="2022-03-28",
     n_parameters=None,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -420,6 +429,7 @@ monot5_3b = ModelMeta(
     revision="bc0c419a438c81f592f878ce32430a1823f5db6c",
     release_date="2022-03-28",
     n_parameters=None,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -476,6 +486,7 @@ flant5_base = ModelMeta(
         # "qed": ["train"],
     ),
     n_parameters=None,
+    n_embedding_parameters=24_674_304,
     memory_usage_mb=944,
     max_tokens=None,
     embed_dim=None,
@@ -522,6 +533,7 @@ flant5_large = ModelMeta(
         # "qed": ["train"],
     ),
     n_parameters=None,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=2987,
     max_tokens=None,
     embed_dim=None,
@@ -568,6 +580,7 @@ flant5_xl = ModelMeta(
         # "qed": ["train"],
     ),
     n_parameters=None,
+    n_embedding_parameters=65_798_144,
     memory_usage_mb=10871,
     max_tokens=None,
     embed_dim=None,
@@ -614,6 +627,7 @@ flant5_xxl = ModelMeta(
         # "qed": ["train"],
     ),
     n_parameters=None,
+    n_embedding_parameters=131_596_288,
     memory_usage_mb=42980,
     max_tokens=None,
     embed_dim=None,
@@ -638,6 +652,7 @@ llama2_7b = ModelMeta(
     revision="01c7f73d771dfac7d292323805ebc428287df4f9",
     release_date="2023-07-18",
     n_parameters=None,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -680,6 +695,7 @@ llama2_7b_chat = ModelMeta(
       url={https://arxiv.org/abs/2307.09288},
     }""",
     n_parameters=None,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -704,6 +720,7 @@ mistral_7b = ModelMeta(
     revision="3ad372fc79158a2148299e3318516c786aeded6c",
     release_date="2023-12-11",
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -740,6 +757,7 @@ followir_7b = ModelMeta(
         # "jhu-clsp/FollowIR-train"
     ),
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=13813,
     max_tokens=None,
     embed_dim=None,
@@ -890,6 +908,7 @@ mt5_base_mmarco_v2 = ModelMeta(
     """,
     training_datasets={"MSMARCO"},
     n_parameters=None,
+    n_embedding_parameters=192_086_016,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,
@@ -913,6 +932,7 @@ mt5_13b_mmarco_100k = ModelMeta(
     revision="e1a4317e102a525ea9e16745ad21394a4f1bffbc",
     release_date="2022-11-04",
     n_parameters=None,
+    n_embedding_parameters=1_024_458_752,
     memory_usage_mb=None,
     max_tokens=None,
     embed_dim=None,

mteb/models/model_implementations/richinfoai_models.py

@@ -15,6 +15,7 @@ ritrieve_zh_v1 = ModelMeta(
     revision="f8d5a707656c55705027678e311f9202c8ced12c",
     release_date="2025-03-25",
     n_parameters=int(326 * 1e6),
+    n_embedding_parameters=21_635_072,
     memory_usage_mb=1242,
     embed_dim=1792,
     license="mit",

mteb/models/model_implementations/ru_sentence_models.py

@@ -244,6 +244,7 @@ rubert_tiny = ModelMeta(
     revision="5441c5ea8026d4f6d7505ec004845409f1259fb1",
     release_date="2021-05-24",
     n_parameters=11_900_000,
+    n_embedding_parameters=9_223_968,
     memory_usage_mb=45,
     embed_dim=312,
     license="mit",
@@ -270,6 +271,7 @@ rubert_tiny2 = ModelMeta(
     revision="dad72b8f77c5eef6995dd3e4691b758ba56b90c3",
     release_date="2021-10-28",
     n_parameters=29_400_000,
+    n_embedding_parameters=26_154_336,
     memory_usage_mb=112,
     embed_dim=312,
     license="mit",
@@ -297,6 +299,7 @@ sbert_large_nlu_ru = ModelMeta(
     revision="af977d5dfa46a3635e29bf0ef383f2df2a08d47a",
     release_date="2020-11-20",
     n_parameters=427_000_000,
+    n_embedding_parameters=123_021_312,
     memory_usage_mb=1629,
     embed_dim=1024,
     license="mit",
@@ -323,6 +326,7 @@ sbert_large_mt_nlu_ru = ModelMeta(
     revision="05300876c2b83f46d3ddd422a7f17e45cf633bb0",
     release_date="2021-05-18",
     n_parameters=427_000_000,
+    n_embedding_parameters=123_021_312,
     memory_usage_mb=1629,
     embed_dim=1024,
     license="not specified",
@@ -351,6 +355,7 @@ user_base_ru = ModelMeta(
     revision="436a489a2087d61aa670b3496a9915f84e46c861",
     release_date="2024-06-10",
     n_parameters=427_000_000,
+    n_embedding_parameters=38_603_520,
     memory_usage_mb=473,
     embed_dim=768,
     license="apache-2.0",
@@ -412,6 +417,7 @@ user_bge_m3 = ModelMeta(
     revision="0cc6cfe48e260fb0474c753087a69369e88709ae",
     release_date="2024-07-05",
     n_parameters=359_026_688,
+    n_embedding_parameters=47_273_984,
     memory_usage_mb=1370,
     embed_dim=1024,
     license="apache-2.0",
@@ -463,6 +469,7 @@ deberta_v1_ru = ModelMeta(
     revision="bdd30b0e19757e6940c92c7aff19e8fc0a60dff4",
     release_date="2023-02-07",
     n_parameters=124_000_000,
+    n_embedding_parameters=38_603_520,
     memory_usage_mb=473,
     embed_dim=768,
     license="apache-2.0",
@@ -494,6 +501,7 @@ rubert_base_cased = ModelMeta(
     revision="4036cab694767a299f2b9e6492909664d9414229",
     release_date="2020-03-04",
     n_parameters=1280_000_000,
+    n_embedding_parameters=91_812_096,
     memory_usage_mb=4883,
     embed_dim=768,
     license="not specified",
@@ -530,6 +538,7 @@ distilrubert_small_cased_conversational = ModelMeta(
     revision="e348066b4a7279b97138038299bddc6580a9169a",
     release_date="2022-06-28",
     n_parameters=107_000_000,
+    n_embedding_parameters=91_812_096,
     memory_usage_mb=408,
     embed_dim=768,
     license="not specified",
@@ -565,6 +574,7 @@ rubert_base_cased_sentence = ModelMeta(
     revision="78b5122d6365337dd4114281b0d08cd1edbb3bc8",
     release_date="2020-03-04",
     n_parameters=107_000_000,
+    n_embedding_parameters=91_812_096,
     memory_usage_mb=408,
     embed_dim=768,
     license="not specified",
@@ -590,6 +600,7 @@ labse_en_ru = ModelMeta(
     revision="cf0714e606d4af551e14ad69a7929cd6b0da7f7e",
     release_date="2021-06-10",
     n_parameters=129_000_000,
+    n_embedding_parameters=42_303_744,
     memory_usage_mb=492,
     embed_dim=768,
     license="not specified",
@@ -618,6 +629,7 @@ rubert_tiny_turbo = ModelMeta(
     revision="8ce0cf757446ce9bb2d5f5a4ac8103c7a1049054",
     release_date="2024-06-21",
     n_parameters=29_200_000,
+    n_embedding_parameters=26_154_336,
     memory_usage_mb=111,
     embed_dim=312,
     license="mit",
@@ -641,6 +653,7 @@ rubert_mini_frida = ModelMeta(
     revision="19b279b78afd945b5ccae78f63e284909814adc2",
     release_date="2025-03-02",
     n_parameters=32_300_000,
+    n_embedding_parameters=26_154_336,
     memory_usage_mb=123,
     embed_dim=312,
     license="mit",
@@ -669,6 +682,7 @@ labse_ru_turbo = ModelMeta(
     revision="1940b046c6b5e125df11722b899130329d0a46da",
     release_date="2024-06-27",
     n_parameters=129_000_000,
+    n_embedding_parameters=42_303_744,
     memory_usage_mb=490,
     embed_dim=768,
     license="mit",
@@ -720,6 +734,7 @@ rosberta_ru_en = ModelMeta(
     use_instructions=True,
     reference="https://huggingface.co/ai-forever/ru-en-RoSBERTa",
     n_parameters=404_000_000,
+    n_embedding_parameters=100_869_120,
     memory_usage_mb=1540,
     max_tokens=512,
     embed_dim=1024,
@@ -886,6 +901,7 @@ frida = ModelMeta(
     use_instructions=True,
     reference="https://huggingface.co/ai-forever/FRIDA",
     n_parameters=823_000_000,
+    n_embedding_parameters=143_847_936,
     memory_usage_mb=3141,
     max_tokens=512,
     embed_dim=1536,
@@ -918,6 +934,7 @@ giga_embeddings = ModelMeta(
     revision="0ad5b29bfecd806cecc9d66b927d828a736594dc",
     release_date="2025-09-23",
     n_parameters=3_227_176_961,
+    n_embedding_parameters=None,
     memory_usage_mb=12865,
     embed_dim=2048,
     license="mit",
@@ -950,6 +967,7 @@ berta = ModelMeta(
     revision="914c8c8aed14042ed890fc2c662d5e9e66b2faa7",
     release_date="2025-03-10",
     n_parameters=128_000_000,
+    n_embedding_parameters=42_303_744,
     memory_usage_mb=489,
     embed_dim=768,
     license="mit",
@@ -1025,6 +1043,7 @@ user2_small = ModelMeta(
     use_instructions=True,
     reference="https://huggingface.co/collections/deepvk/user2-6802650d7210f222ec60e05f",
     n_parameters=34_400_000,
+    n_embedding_parameters=None,
     memory_usage_mb=131,
     max_tokens=8192,
     embed_dim=384,
@@ -1058,6 +1077,7 @@ user2_base = ModelMeta(
     use_instructions=True,
     reference="https://huggingface.co/collections/deepvk/user2-6802650d7210f222ec60e05f",
     n_parameters=149_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=568,
     max_tokens=8192,
     embed_dim=768,

mteb/models/model_implementations/ruri_models.py

@@ -38,6 +38,7 @@ cl_nagoya_ruri_v3_30m = ModelMeta(
     revision="24899e5de370b56d179604a007c0d727bf144504",
     release_date="2025-04-07",
     n_parameters=36_705_536,
+    n_embedding_parameters=None,
     memory_usage_mb=140,
     embed_dim=256,
     license="apache-2.0",
@@ -69,6 +70,7 @@ cl_nagoya_ruri_v3_70m = ModelMeta(
     revision="07a8b0aba47d29d2ca21f89b915c1efe2c23d1cc",
     release_date="2025-04-09",
     n_parameters=36_705_536,
+    n_embedding_parameters=None,
     memory_usage_mb=140,
     embed_dim=256,
     license="apache-2.0",
@@ -98,6 +100,7 @@ cl_nagoya_ruri_v3_130m = ModelMeta(
     revision="e3114c6ee10dbab8b4b235fbc6dcf9dd4d5ac1a6",
     release_date="2025-04-09",
     n_parameters=132_140_544,
+    n_embedding_parameters=None,
     memory_usage_mb=504,
     embed_dim=512,
     license="apache-2.0",
@@ -127,6 +130,7 @@ cl_nagoya_ruri_v3_310m = ModelMeta(
     revision="18b60fb8c2b9df296fb4212bb7d23ef94e579cd3",
     release_date="2025-04-09",
     n_parameters=314_611_968,
+    n_embedding_parameters=None,
     memory_usage_mb=1200,
     embed_dim=768,
     license="apache-2.0",
@@ -157,6 +161,7 @@ cl_nagoya_ruri_small_v2 = ModelMeta(
     revision="db18646e673b713cd0518a5bb0fefdce21e77cd9",
     release_date="2024-12-05",
     n_parameters=68_087_808,
+    n_embedding_parameters=25_165_824,
     memory_usage_mb=260,
     embed_dim=768,
     license="apache-2.0",
@@ -186,6 +191,7 @@ cl_nagoya_ruri_base_v2 = ModelMeta(
     revision="8ce03882903668a01c83ca3b8111ac025a3bc734",
     release_date="2024-12-05",
     n_parameters=111_207_168,
+    n_embedding_parameters=25_165_824,
     memory_usage_mb=424,
     embed_dim=768,
     license="apache-2.0",
@@ -215,6 +221,7 @@ cl_nagoya_ruri_large_v2 = ModelMeta(
     revision="42898ef34a5574977380ebf0dfd28cbfbd36438b",
     release_date="2024-12-06",
     n_parameters=337_441_792,
+    n_embedding_parameters=33_554_432,
     memory_usage_mb=1287,
     embed_dim=1024,
     license="apache-2.0",
@@ -245,6 +252,7 @@ cl_nagoya_ruri_small_v1 = ModelMeta(
     revision="bc56ce90cd7a979f6eb199fc52dfe700bfd94bc3",
     release_date="2024-08-28",
     n_parameters=68_087_808,
+    n_embedding_parameters=25_165_824,
     memory_usage_mb=130,
     embed_dim=768,
     license="apache-2.0",
@@ -274,6 +282,7 @@ cl_nagoya_ruri_base_v1 = ModelMeta(
     revision="1ae40b8b6c78518a499425086bab8fc16c2e4b0e",
     release_date="2024-08-28",
     n_parameters=111_207_168,
+    n_embedding_parameters=25_165_824,
     memory_usage_mb=212,
     embed_dim=768,
     license="apache-2.0",
@@ -304,6 +313,7 @@ cl_nagoya_ruri_large_v1 = ModelMeta(
     revision="a011c39b13e8bc137ee13c6bc82191ece46c414c",
     release_date="2024-08-28",
     n_parameters=337_441_792,
+    n_embedding_parameters=33_554_432,
     memory_usage_mb=644,
     embed_dim=1024,
     license="apache-2.0",

mteb/models/model_implementations/salesforce_models.py

@@ -1,12 +1,18 @@
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
 from mteb.models.instruct_wrapper import (
     InstructSentenceTransformerModel,
     instruct_wrapper,
 )
 from mteb.models.model_meta import ModelMeta, ScoringFunction
-from mteb.types import PromptType
 
 from .e5_instruct import E5_MISTRAL_TRAINING_DATA
 
+if TYPE_CHECKING:
+    from mteb.types import PromptType
+
 
 def instruction_template(
     instruction: str, prompt_type: PromptType | None = None
@@ -52,6 +58,7 @@ SFR_Embedding_2_R = ModelMeta(
     revision="91762139d94ed4371a9fa31db5551272e0b83818",
     release_date="2024-06-14",  # initial commit of hf model.
     n_parameters=7_110_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=13563,
     embed_dim=4096,
     license="cc-by-nc-4.0",
@@ -90,6 +97,7 @@ SFR_Embedding_Code_2B_R = ModelMeta(
     revision="c73d8631a005876ed5abde34db514b1fb6566973",
     release_date="2025-01-17",  # initial commit of hf model.
     n_parameters=2_610_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=4986,
     embed_dim=2304,
     license="cc-by-nc-4.0",
@@ -128,6 +136,7 @@ SFR_Embedding_Mistral = ModelMeta(
     revision="938c560d1c236aa563b2dbdf084f28ab28bccb11",
     release_date="2024-01-24",  # initial commit of hf model.
     n_parameters=7_110_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=13563,
     embed_dim=4096,
     license="cc-by-nc-4.0",

mteb/models/model_implementations/samilpwc_models.py

@@ -51,6 +51,7 @@ samilpwc_expr = ModelMeta(
     revision="33358978be40f36491045f9c2a359d38c3f50047",
     release_date="2025-08-12",
     n_parameters=560_000_000,
+    n_embedding_parameters=256_002_048,
     memory_usage_mb=2136,
     embed_dim=1024,
     license="apache-2.0",

mteb/models/model_implementations/sarashina_embedding_models.py

@@ -124,6 +124,7 @@ sbintuitions_sarashina_embedding_v2_1b = ModelMeta(
     revision="1f3408afaa7b617e3445d891310a9c26dd0c68a5",
     release_date="2025-07-30",
     n_parameters=1_224_038_144,
+    n_embedding_parameters=183_500_800,
     memory_usage_mb=4669,
     embed_dim=1792,
     license="https://huggingface.co/sbintuitions/sarashina-embedding-v2-1b/blob/main/LICENSE",
@@ -150,6 +151,7 @@ sbintuitions_sarashina_embedding_v1_1b = ModelMeta(
     revision="d060fcd8984075071e7fad81baff035cbb3b6c7e",
     release_date="2024-11-22",
     n_parameters=1_224_038_144,
+    n_embedding_parameters=183_500_800,
     memory_usage_mb=4669,
     embed_dim=1792,
     license="https://huggingface.co/sbintuitions/sarashina-embedding-v1-1b/blob/main/LICENSE",

mteb/models/model_implementations/searchmap_models.py

@@ -27,6 +27,7 @@ searchmap_preview = ModelMeta(
     use_instructions=True,
     release_date="2025-03-05",
     n_parameters=435_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=1660,
     embed_dim=4096,
     license="mit",

mteb/models/model_implementations/seed_1_6_embedding_models.py

@@ -13,16 +13,18 @@ import torch
 from torch.utils.data import DataLoader
 
 from mteb._requires_package import requires_package
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_implementations.bge_models import bge_chinese_training_data
 from mteb.models.model_implementations.nvidia_models import nvidia_training_datasets
 from mteb.models.model_meta import ModelMeta
-from mteb.types import Array, BatchedInput, PromptType
+from mteb.types import PromptType
 
 if TYPE_CHECKING:
     from PIL import Image
 
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput
+
 
 logger = logging.getLogger(__name__)
 
@@ -429,6 +431,7 @@ seed_embedding = ModelMeta(
     embed_dim=2048,
     open_weights=False,
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     license=None,
     reference="https://seed1-6-embedding.github.io/",

mteb/models/model_implementations/seed_1_6_embedding_models_1215.py

@@ -15,15 +15,18 @@ from torch.utils.data import DataLoader
 from tqdm import tqdm
 
 from mteb._requires_package import requires_package
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_implementations.bge_models import bge_chinese_training_data
 from mteb.models.model_implementations.nvidia_models import nvidia_training_datasets
 from mteb.models.model_meta import ModelMeta
-from mteb.types import Array, BatchedInput, PromptType
+from mteb.types import PromptType
 
 if TYPE_CHECKING:
     from PIL import Image
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput
 
 
 logger = logging.getLogger(__name__)
@@ -613,6 +616,7 @@ seed_embedding = ModelMeta(
     embed_dim=2048,
     open_weights=False,
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     license=None,
     reference="https://console.volcengine.com/ark/region:ark+cn-beijing/model/detail?Id=doubao-embedding-vision",

mteb/models/model_implementations/seed_models.py

@@ -9,7 +9,7 @@ from tqdm.auto import tqdm
 from mteb._requires_package import requires_package
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_meta import ModelMeta
-from mteb.models.models_protocols import PromptType
+from mteb.types import PromptType
 
 from .bge_models import bge_chinese_training_data
 from .nvidia_models import nvidia_training_datasets
@@ -253,6 +253,7 @@ seed_embedding = ModelMeta(
     embed_dim=2048,
     open_weights=False,
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     license=None,
     reference="https://seed1-5-embedding.github.io/",

mteb/models/model_implementations/sentence_transformers_models.py

@@ -121,6 +121,7 @@ all_minilm_l6_v2 = ModelMeta(
     revision="8b3219a92973c328a8e22fadcfa821b5dc75636a",
     release_date="2021-08-30",
     n_parameters=22_700_000,
+    n_embedding_parameters=11_720_448,
     memory_usage_mb=87,
     embed_dim=384,
     license="apache-2.0",
@@ -152,6 +153,7 @@ all_minilm_l12_v2 = ModelMeta(
     revision="364dd28d28dcd3359b537f3cf1f5348ba679da62",
     release_date="2021-08-30",
     n_parameters=33_400_000,
+    n_embedding_parameters=11_720_448,
     memory_usage_mb=127,
     embed_dim=384,
     license="apache-2.0",
@@ -183,6 +185,7 @@ paraphrase_multilingual_minilm_l12_v2 = ModelMeta(
     revision="bf3bf13ab40c3157080a7ab344c831b9ad18b5eb",
     release_date="2019-11-01",  # release date of paper
     n_parameters=118_000_000,
+    n_embedding_parameters=96_014_208,
     memory_usage_mb=449,
     embed_dim=768,
     license="apache-2.0",
@@ -214,6 +217,7 @@ paraphrase_multilingual_mpnet_base_v2 = ModelMeta(
     revision="79f2382ceacceacdf38563d7c5d16b9ff8d725d6",
     release_date="2019-11-01",  # release date of paper
     n_parameters=278_000_000,
+    n_embedding_parameters=192_001_536,
     memory_usage_mb=1061,
     embed_dim=768,
     license="apache-2.0",
@@ -256,6 +260,7 @@ labse = ModelMeta(
     revision="e34fab64a3011d2176c99545a93d5cbddc9a91b7",
     release_date="2019-11-01",  # release date of paper
     n_parameters=471_000_000,
+    n_embedding_parameters=384_885_504,
     memory_usage_mb=1796,
     embed_dim=768,
     license="apache-2.0",
@@ -294,6 +299,7 @@ multi_qa_minilm_l6_cos_v1 = ModelMeta(
     revision="b207367332321f8e44f96e224ef15bc607f4dbf0",
     release_date="2021-08-30",
     n_parameters=22_700_000,
+    n_embedding_parameters=11_720_448,
     memory_usage_mb=87,
     embed_dim=384,
     license="apache-2.0",
@@ -325,6 +331,7 @@ all_mpnet_base_v2 = ModelMeta(
     revision="9a3225965996d404b775526de6dbfe85d3368642",
     release_date="2021-08-30",
     n_parameters=109_000_000,
+    n_embedding_parameters=23_444_736,
     memory_usage_mb=418,
     embed_dim=768,
     license="apache-2.0",
@@ -435,6 +442,7 @@ static_similarity_mrl_multilingual_v1 = ModelMeta(
     revision="7264ea07c5365a11d7e6d87dbb6195889a13054f",
     release_date="2025-01-15",
     n_parameters=108_420_096,
+    n_embedding_parameters=None,
     memory_usage_mb=413,
     embed_dim=1024,
     license="apache-2.0",
@@ -468,6 +476,7 @@ contriever = ModelMeta(
     revision="abe8c1493371369031bcb1e02acb754cf4e162fa",
     release_date="2022-06-25",  # release date of model on HF
     n_parameters=150_000_000,
+    n_embedding_parameters=23_440_896,
     memory_usage_mb=572,
     embed_dim=768,
     license=None,
@@ -498,6 +507,7 @@ microllama_text_embedding = ModelMeta(
     revision="98f70f14cdf12d7ea217ed2fd4e808b0195f1e7e",
     release_date="2024-11-10",
     n_parameters=272_000_000,
+    n_embedding_parameters=32_769_024,
     memory_usage_mb=1037,
     embed_dim=1024,
     license="apache-2.0",
@@ -544,6 +554,7 @@ sentence_t5_base = ModelMeta(
     revision="50c53e206f8b01c9621484a3c0aafce4e55efebf",
     release_date="2022-02-09",
     n_parameters=110_000_000,
+    n_embedding_parameters=24_674_304,
     memory_usage_mb=209,
     embed_dim=768,
     license="apache-2.0",
@@ -567,6 +578,7 @@ sentence_t5_large = ModelMeta(
     revision="1fc08ea477205aa54a3e5b13f0971ae16b86410a",
     release_date="2022-02-09",
     n_parameters=335_000_000,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=639,
     embed_dim=768,
     license="apache-2.0",
@@ -590,6 +602,7 @@ sentence_t5_xl = ModelMeta(
     revision="2965d31b368fb14117688e0bde77cbd720e91f53",
     release_date="2024-03-27",
     n_parameters=3_000_000_000,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=2367,
     embed_dim=768,
     license="apache-2.0",
@@ -613,6 +626,7 @@ sentence_t5_xxl = ModelMeta(
     revision="4d122282ba80e807e9e6eb8c358269e92796365d",
     release_date="2024-03-27",
     n_parameters=11_000_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=9279,
     embed_dim=768,
     license="apache-2.0",
@@ -646,6 +660,7 @@ gtr_t5_large = ModelMeta(
     revision="a2c8ac47f998531948d4cbe32a0b577a7037a5e3",
     release_date="2022-02-09",
     n_parameters=335_000_000,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=639,
     embed_dim=768,
     license="apache-2.0",
@@ -681,6 +696,7 @@ gtr_t5_xl = ModelMeta(
     revision="23a8d667a1ad2578af181ce762867003c498d1bf",
     release_date="2022-02-09",
     n_parameters=1_240_000_000,
+    n_embedding_parameters=32_899_072,
     memory_usage_mb=2367,
     embed_dim=768,
     license="apache-2.0",
@@ -715,6 +731,7 @@ gtr_t5_xxl = ModelMeta(
     revision="73f2a9156a3dcc2194dfdb2bf201cd7d17e17884",
     release_date="2022-02-09",
     n_parameters=4_860_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=9279,
     embed_dim=768,
     license="apache-2.0",
@@ -750,6 +767,7 @@ gtr_t5_base = ModelMeta(
     revision="7027e9594267928589816394bdd295273ddc0739",
     release_date="2022-02-09",
     n_parameters=110_000_000,
+    n_embedding_parameters=24_674_304,
     memory_usage_mb=209,
     embed_dim=768,
     license="apache-2.0",

mteb/models/model_implementations/shuu_model.py

@@ -10,6 +10,7 @@ codemodernbert_crow_meta = ModelMeta(
     revision="044a7a4b552f86e284817234c336bccf16f895ce",
     release_date="2025-04-21",
     n_parameters=151668480,
+    n_embedding_parameters=None,
     memory_usage_mb=607,
     embed_dim=768,
     license="apache-2.0",