mteb 2.7.2__py3-none-any.whl → 2.7.4__py3-none-any.whl

mteb/models/model_implementations/facebookai.py

@@ -113,6 +113,7 @@ xlmr_base = ModelMeta(
     revision="e73636d4f797dec63c3081bb6ed5c7b0bb3f2089",
     release_date="2019-11-05",  # arxiv paper release
     n_parameters=278043648,
+    n_embedding_parameters=192_001_536,
     memory_usage_mb=1064,
     embed_dim=768,
     license="mit",
@@ -163,6 +164,7 @@ xlmr_large = ModelMeta(
     revision="c23d21b0620b635a76227c604d44e43a9f0ee389",
     release_date="2019-11-05",  # arxiv paper release
     n_parameters=559890432,
+    n_embedding_parameters=256_002_048,
     memory_usage_mb=2141,
     embed_dim=1024,
     license="mit",

mteb/models/model_implementations/geogpt_models.py

@@ -20,6 +20,7 @@ geoembedding = ModelMeta(
     ),
     release_date="2025-04-22",
     n_parameters=7241732096,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=27625,
     embed_dim=4096,
     license="apache-2.0",

mteb/models/model_implementations/gme_v_models.py

@@ -6,16 +6,18 @@ import warnings
 from typing import TYPE_CHECKING, Any
 
 import torch
-from torch.utils.data import DataLoader
 from tqdm.autonotebook 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
+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__)
 
@@ -354,6 +356,7 @@ gme_qwen2vl_2b = ModelMeta(
     release_date="2024-12-24",
     modalities=["image", "text"],
     n_parameters=2_210_000_000,
+    n_embedding_parameters=233_373_696,
     memory_usage_mb=8427,
     embed_dim=1536,
     license="apache-2.0",
@@ -378,6 +381,7 @@ gme_qwen2vl_7b = ModelMeta(
     release_date="2024-12-24",
     modalities=["image", "text"],
     n_parameters=8_290_000_000,
+    n_embedding_parameters=544_997_376,
     memory_usage_mb=31629,
     embed_dim=3584,
     license="apache-2.0",

mteb/models/model_implementations/google_models.py

@@ -1,17 +1,23 @@
-from typing import Any
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any
 
 import numpy as np
 from packaging.version import Version
-from torch.utils.data import DataLoader
 from tqdm.auto import tqdm
 from transformers import __version__ as transformers_version
 
 from mteb._requires_package import requires_package
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models import sentence_transformers_loader
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_meta import ModelMeta, ScoringFunction
-from mteb.types import Array, BatchedInput, PromptType
+from mteb.types import PromptType
+
+if TYPE_CHECKING:
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput
 
 MULTILINGUAL_EVALUATED_LANGUAGES = [
     "arb-Arab",
@@ -156,6 +162,7 @@ google_text_emb_004 = ModelMeta(
     revision="1",  # revision is intended for implementation
     release_date="2024-05-14",
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     max_tokens=2048,
     embed_dim=768,
@@ -181,6 +188,7 @@ google_text_emb_005 = ModelMeta(
     revision="1",  # revision is intended for implementation
     release_date="2024-11-18",
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     max_tokens=2048,
     embed_dim=768,
@@ -206,6 +214,7 @@ google_text_multilingual_emb_002 = ModelMeta(
     revision="1",
     release_date="2024-05-14",
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     max_tokens=2048,
     embed_dim=768,
@@ -231,6 +240,7 @@ google_gemini_embedding_001 = ModelMeta(
     revision="1",
     release_date="2025-03-07",
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     max_tokens=2048,
     embed_dim=3072,
@@ -266,6 +276,7 @@ embedding_gemma_300m = ModelMeta(
     revision="64614b0b8b64f0c6c1e52b07e4e9a4e8fe4d2da2",
     release_date="2025-09-04",
     n_parameters=307_581_696,
+    n_embedding_parameters=201_326_592,
     embed_dim=768,
     max_tokens=2048,
     license="gemma",

mteb/models/model_implementations/granite_vision_embedding_models.py

@@ -4,20 +4,21 @@ import logging
 from typing import TYPE_CHECKING, Any
 
 import torch
-from torch.utils.data import DataLoader
 from tqdm.auto import tqdm
 
 from mteb._requires_package import (
     requires_image_dependencies,
 )
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.model_meta import ModelMeta
-from mteb.types import Array, BatchedInput, PromptType
-
-logger = logging.getLogger(__name__)
 
 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, PromptType
+
+logger = logging.getLogger(__name__)
 
 
 class GraniteVisionEmbeddingWrapper:
@@ -172,6 +173,7 @@ granite_vision_embedding = ModelMeta(
     release_date="2025-06-11",
     modalities=["image", "text"],
     n_parameters=2_980_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=11351,
     max_tokens=128000,
     embed_dim=128,

mteb/models/model_implementations/gritlm_models.py

@@ -44,6 +44,7 @@ gritlm7b = ModelMeta(
     revision="13f00a0e36500c80ce12870ea513846a066004af",
     release_date="2024-02-15",
     n_parameters=7_240_000_000,
+    n_embedding_parameters=131_072_000,
     memory_usage_mb=13813,
     embed_dim=4096,
     license="apache-2.0",
@@ -73,6 +74,7 @@ gritlm8x7b = ModelMeta(
     revision="7f089b13e3345510281733ca1e6ff871b5b4bc76",
     release_date="2024-02-15",
     n_parameters=57_920_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=89079,
     embed_dim=32768,
     license="apache-2.0",

mteb/models/model_implementations/gte_models.py

@@ -48,6 +48,7 @@ gte_qwen2_7b_instruct = ModelMeta(
     revision="e26182b2122f4435e8b3ebecbf363990f409b45b",
     release_date="2024-06-15",  # initial commit of hf model.
     n_parameters=7_613_000_000,
+    n_embedding_parameters=543_499_264,
     memory_usage_mb=29040,
     embed_dim=3584,
     license="apache-2.0",
@@ -80,6 +81,7 @@ gte_qwen1_5_7b_instruct = ModelMeta(
     revision="07d27e5226328010336563bc1b564a5e3436a298",
     release_date="2024-04-20",  # initial commit of hf model.
     n_parameters=7_720_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=29449,
     embed_dim=4096,
     license="apache-2.0",
@@ -117,6 +119,7 @@ gte_qwen2_1_5b_instruct = ModelMeta(
     revision="c6c1b92f4a3e1b92b326ad29dd3c8433457df8dd",
     release_date="2024-07-29",  # initial commit of hf model.
     n_parameters=1_780_000_000,
+    n_embedding_parameters=232_928_256,
     memory_usage_mb=6776,
     embed_dim=8960,
     license="apache-2.0",
@@ -145,6 +148,7 @@ gte_small_zh = ModelMeta(
     revision="af7bd46fbb00b3a6963c8dd7f1786ddfbfbe973a",
     release_date="2023-11-08",  # initial commit of hf model.
     n_parameters=int(30.3 * 1e6),
+    n_embedding_parameters=10_817_536,
     memory_usage_mb=58,
     embed_dim=1024,
     license="mit",
@@ -173,6 +177,7 @@ gte_base_zh = ModelMeta(
     revision="71ab7947d6fac5b64aa299e6e40e6c2b2e85976c",
     release_date="2023-11-08",  # initial commit of hf model.
     n_parameters=int(102 * 1e6),
+    n_embedding_parameters=16_226_304,
     memory_usage_mb=195,
     embed_dim=1024,
     license="mit",
@@ -201,6 +206,7 @@ gte_large_zh = ModelMeta(
     revision="64c364e579de308104a9b2c170ca009502f4f545",
     release_date="2023-11-08",  # initial commit of hf model.
     n_parameters=int(326 * 1e6),
+    n_embedding_parameters=21_635_072,
     memory_usage_mb=621,
     embed_dim=1024,
     license="mit",
@@ -330,6 +336,7 @@ gte_multilingual_base = ModelMeta(
     revision="ca1791e0bcc104f6db161f27de1340241b13c5a4",
     release_date="2024-07-20",  # initial commit of hf model.
     n_parameters=int(305 * 1e6),
+    n_embedding_parameters=192_036_864,
     memory_usage_mb=582,
     embed_dim=768,
     license="apache-2.0",
@@ -359,6 +366,7 @@ gte_modernbert_base = ModelMeta(
     revision="7ca8b4ca700621b67618669f5378fe5f5820b8e4",
     release_date="2025-01-21",  # initial commit of hf model.
     n_parameters=int(149 * 1e6),
+    n_embedding_parameters=None,
     memory_usage_mb=284,
     embed_dim=768,
     license="apache-2.0",
@@ -402,6 +410,7 @@ gte_base_en_v15 = ModelMeta(
     revision="a829fd0e060bb84554da0dfd354d0de0f7712b7f",  # can be any
     release_date="2024-06-20",  # initial commit of hf model
     n_parameters=137_000_000,
+    n_embedding_parameters=23_445_504,
     memory_usage_mb=None,
     embed_dim=768,
     license="apache-2.0",

mteb/models/model_implementations/hinvec_models.py

@@ -1,9 +1,13 @@
+from __future__ import annotations
+
 import logging
+from typing import TYPE_CHECKING
 
 from mteb.models.model_meta import ModelMeta
 from mteb.models.sentence_transformer_wrapper import sentence_transformers_loader
-from mteb.types import PromptType
 
+if TYPE_CHECKING:
+    from mteb.types import PromptType
 logger = logging.getLogger(__name__)
 
 
@@ -43,6 +47,7 @@ Hinvec_bidir = ModelMeta(
     revision="d4fc678720cc1b8c5d18599ce2d9a4d6090c8b6b",
     release_date="2025-06-19",
     n_parameters=939_591_680,
+    n_embedding_parameters=None,
     memory_usage_mb=3715,
     embed_dim=2048,
     license="cc-by-nc-4.0",

mteb/models/model_implementations/human.py

@@ -9,6 +9,7 @@ human = ModelMeta(
     revision="2025_09_25",
     release_date=None,
     n_parameters=None,
+    n_embedding_parameters=None,
     memory_usage_mb=None,
     embed_dim=None,
     license=None,

mteb/models/model_implementations/ibm_granite_models.py

@@ -100,6 +100,7 @@ granite_107m_multilingual = ModelMeta(
     revision="47db56afe692f731540413c67dd818ff492277e7",
     release_date="2024-12-18",
     n_parameters=107_000_000,
+    n_embedding_parameters=96_000_768,
     memory_usage_mb=204,
     embed_dim=384,
     license="apache-2.0",
@@ -131,6 +132,7 @@ granite_278m_multilingual = ModelMeta(
     revision="84e3546b88b0cb69f8078608a1df558020bcbf1f",
     release_date="2024-12-18",
     n_parameters=278_000_000,
+    n_embedding_parameters=192_001_536,
     memory_usage_mb=530,
     embed_dim=768,
     license="apache-2.0",
@@ -162,6 +164,7 @@ granite_30m_english = ModelMeta(
     revision="eddbb57470f896b5f8e2bfcb823d8f0e2d2024a5",
     release_date="2024-12-18",
     n_parameters=30_000_000,
+    n_embedding_parameters=19_301_760,
     memory_usage_mb=58,
     embed_dim=384,
     license="apache-2.0",
@@ -193,6 +196,7 @@ granite_125m_english = ModelMeta(
     revision="e48d3a5b47eaa18e3fe07d4676e187fd80f32730",
     release_date="2024-12-18",
     n_parameters=125_000_000,
+    n_embedding_parameters=38_603_520,
     memory_usage_mb=238,
     embed_dim=768,
     license="apache-2.0",
@@ -225,6 +229,7 @@ granite_english_r2 = ModelMeta(
     revision="6e7b8ce0e76270394ac4669ba4bbd7133b60b7f9",
     release_date="2025-08-15",
     n_parameters=149_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=284,
     embed_dim=768,
     license="apache-2.0",
@@ -250,6 +255,7 @@ granite_small_english_r2 = ModelMeta(
     revision="54a8d2616a0844355a5164432d3f6dafb37b17a3",
     release_date="2025-08-15",
     n_parameters=47_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=91,
     embed_dim=384,
     license="apache-2.0",

mteb/models/model_implementations/inf_models.py

@@ -56,6 +56,7 @@ inf_retriever_v1 = ModelMeta(
     revision="cb70ca7c31dfa866b2eff2dad229c144d8ddfd91",
     release_date="2024-12-24",  # initial commit of hf model.
     n_parameters=7_069_121_024,
+    n_embedding_parameters=None,
     memory_usage_mb=13483,
     embed_dim=3584,
     license="apache-2.0",
@@ -83,6 +84,7 @@ inf_retriever_v1_1_5b = ModelMeta(
     revision="c9c05c2dd50707a486966ba81703021ae2094a06",
     release_date="2025-02-08",  # initial commit of hf model.
     n_parameters=1_543_268_864,
+    n_embedding_parameters=232_928_256,
     memory_usage_mb=2944,
     embed_dim=1536,
     license="apache-2.0",

mteb/models/model_implementations/jasper_models.py

@@ -1,11 +1,10 @@
+from __future__ import annotations
+
 import logging
-from collections.abc import Callable
-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.abs_encoder import AbsEncoder
 from mteb.models.instruct_wrapper import InstructSentenceTransformerModel
 from mteb.models.model_implementations.bge_models import (
@@ -17,7 +16,15 @@ from mteb.models.model_implementations.e5_instruct import E5_MISTRAL_TRAINING_DA
 from mteb.models.model_implementations.nvidia_models import nvidia_training_datasets
 from mteb.models.model_implementations.qzhou_models import qzhou_training_data
 from mteb.models.model_meta import ModelMeta, ScoringFunction
-from mteb.types import Array, BatchedInput, PromptType
+from mteb.types import PromptType
+
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput
 
 logger = logging.getLogger(__name__)
 
@@ -292,6 +299,7 @@ jasper_en_v1 = ModelMeta(
     revision="d6330ce98f8a0d741e781df845904c9484f00efa",
     release_date="2024-12-11",  # first commit
     n_parameters=1_999_000_000,
+    n_embedding_parameters=232_932_864,
     memory_usage_mb=3802,
     max_tokens=131072,
     embed_dim=8960,
@@ -339,6 +347,7 @@ Jasper_Token_Compression_600M = ModelMeta(
     revision="06a100f753a5a96d9e583b3af79c6fcdfacc4719",
     release_date="2025-11-14",
     n_parameters=595776512,
+    n_embedding_parameters=None,
     memory_usage_mb=2272,
     embed_dim=2048,
     license="mit",

mteb/models/model_implementations/jina_clip.py

@@ -1,15 +1,20 @@
-from typing import Any
+from __future__ import annotations
+
+from typing import TYPE_CHECKING, Any
 
 import torch
-from torch.utils.data import DataLoader
 from tqdm.auto import tqdm
 
 from mteb._requires_package import requires_image_dependencies
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_implementations.colpali_models import COLPALI_TRAINING_DATA
 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
 
 JINA_CLIP_CITATION = """@article{koukounas2024jinaclip,
   title={Jina CLIP: Your CLIP Model Is Also Your Text Retriever},
@@ -139,6 +144,7 @@ jina_clip_v1 = ModelMeta(
     release_date="2024-05-30",
     modalities=["image", "text"],
     n_parameters=223_000_000,
+    n_embedding_parameters=None,
     memory_usage_mb=849,
     max_tokens=8192,
     embed_dim=768,

mteb/models/model_implementations/jina_models.py

@@ -1,14 +1,13 @@
+from __future__ import annotations
+
 import logging
 from collections import defaultdict
-from typing import Any, ClassVar
+from typing import TYPE_CHECKING, Any, ClassVar
 
 import numpy as np
 import torch
-from sentence_transformers import CrossEncoder
-from torch.utils.data import DataLoader
 
 from mteb._requires_package import requires_package
-from mteb.abstasks.task_metadata import TaskMetadata
 from mteb.languages import PROGRAMMING_LANGS
 from mteb.models.abs_encoder import AbsEncoder
 from mteb.models.model_meta import ModelMeta, ScoringFunction
@@ -16,7 +15,13 @@ from mteb.models.sentence_transformer_wrapper import (
     CrossEncoderWrapper,
     SentenceTransformerEncoderWrapper,
 )
-from mteb.types import Array, BatchedInput, PromptType
+
+if TYPE_CHECKING:
+    from sentence_transformers import CrossEncoder
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput, PromptType
 
 logger = logging.getLogger(__name__)
 
@@ -728,6 +733,7 @@ jina_reranker_v3 = ModelMeta(
     release_date="2025-09-18",  # official release date
     modalities=["text"],
     n_parameters=int(0.6 * 1e9),
+    n_embedding_parameters=None,
     memory_usage_mb=1138,
     max_tokens=131072,
     embed_dim=None,
@@ -771,6 +777,7 @@ jina_embeddings_v4 = ModelMeta(
     release_date="2025-06-24",  # official release date
     modalities=["image", "text"],
     n_parameters=int(3.8 * 1e9),
+    n_embedding_parameters=None,
     memory_usage_mb=7500,
     max_tokens=32768,
     embed_dim=2048,
@@ -819,6 +826,7 @@ jina_embeddings_v3 = ModelMeta(
     revision="215a6e121fa0183376388ac6b1ae230326bfeaed",
     release_date="2024-09-18",  # official release date
     n_parameters=int(572 * 1e6),
+    n_embedding_parameters=None,
     memory_usage_mb=1092,
     max_tokens=8194,
     embed_dim=1024,
@@ -879,6 +887,7 @@ jina_embeddings_v2_base_en = ModelMeta(
     revision="6e85f575bc273f1fd840a658067d0157933c83f0",
     release_date="2023-09-27",
     n_parameters=137_000_000,
+    n_embedding_parameters=23_445_504,
     memory_usage_mb=262,
     embed_dim=768,
     license="apache-2.0",
@@ -943,6 +952,7 @@ jina_embeddings_v2_small_en = ModelMeta(
     revision="44e7d1d6caec8c883c2d4b207588504d519788d0",
     release_date="2023-09-27",
     n_parameters=32_700_000,
+    n_embedding_parameters=15_630_336,
     memory_usage_mb=62,
     embed_dim=512,
     license="apache-2.0",
@@ -1004,6 +1014,7 @@ jina_embedding_b_en_v1 = ModelMeta(
     revision="32aa658e5ceb90793454d22a57d8e3a14e699516",
     release_date="2023-07-07",
     n_parameters=110_000_000,
+    n_embedding_parameters=24_674_304,
     memory_usage_mb=420,
     embed_dim=768,
     license="apache-2.0",
@@ -1061,6 +1072,7 @@ jina_embedding_s_en_v1 = ModelMeta(
     revision="5ac6cd473e2324c6d5f9e558a6a9f65abb57143e",
     release_date="2023-07-07",
     n_parameters=35_000_000,
+    n_embedding_parameters=16_449_536,
     memory_usage_mb=134,
     embed_dim=512,
     license="apache-2.0",

mteb/models/model_implementations/kalm_models.py

@@ -1,14 +1,20 @@
+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.instruct_wrapper import InstructSentenceTransformerModel
 from mteb.models.model_meta import ModelMeta
 from mteb.models.sentence_transformer_wrapper import sentence_transformers_loader
-from mteb.types import Array, BatchedInput, PromptType
+from mteb.types import PromptType
+
+if TYPE_CHECKING:
+    from torch.utils.data import DataLoader
+
+    from mteb.abstasks.task_metadata import TaskMetadata
+    from mteb.types import Array, BatchedInput
 
 logger = logging.getLogger(__name__)
 
@@ -774,6 +780,7 @@ HIT_TMG__KaLM_embedding_multilingual_mini_instruct_v1 = ModelMeta(
     release_date="2024-10-23",
     languages=["eng-Latn", "zho-Hans"],
     n_parameters=494032768,
+    n_embedding_parameters=136_134_656,
     memory_usage_mb=1885,
     max_tokens=512,
     embed_dim=896,
@@ -799,6 +806,7 @@ HIT_TMG__KaLM_embedding_multilingual_mini_v1 = ModelMeta(
     release_date="2024-08-27",
     languages=["eng-Latn", "zho-Hans"],
     n_parameters=494032768,
+    n_embedding_parameters=136_134_656,
     memory_usage_mb=1885,
     max_tokens=512,
     embed_dim=896,
@@ -830,6 +838,7 @@ HIT_TMG__KaLM_embedding_multilingual_mini_instruct_v1_5 = ModelMeta(
     release_date="2024-12-26",
     languages=["eng-Latn", "zho-Hans"],
     n_parameters=494032768,
+    n_embedding_parameters=136_134_656,
     memory_usage_mb=1885,
     max_tokens=512,
     embed_dim=896,
@@ -861,6 +870,7 @@ HIT_TMG__KaLM_embedding_multilingual_mini_instruct_v2 = ModelMeta(
     release_date="2025-06-25",
     languages=["eng-Latn", "zho-Hans"],
     n_parameters=494032768,
+    n_embedding_parameters=136_134_656,
     memory_usage_mb=942,
     max_tokens=512,
     embed_dim=896,
@@ -892,6 +902,7 @@ KaLM_Embedding_KaLM_embedding_multilingual_mini_instruct_v2_5 = ModelMeta(
     release_date="2025-09-30",
     languages=["eng-Latn", "zho-Hans"],
     n_parameters=494032768,
+    n_embedding_parameters=136_134_656,
     memory_usage_mb=1885,
     max_tokens=512,
     embed_dim=896,
@@ -907,23 +918,23 @@ KaLM_Embedding_KaLM_embedding_multilingual_mini_instruct_v2_5 = ModelMeta(
     adapted_from="HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v2",
     superseded_by=None,
     citation="""@misc{zhao2025kalmembeddingv2,
-      title={KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model}, 
+      title={KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model},
       author={Xinping Zhao and Xinshuo Hu and Zifei Shan and Shouzheng Huang and Yao Zhou and Xin Zhang and Zetian Sun and Zhenyu Liu and Dongfang Li and Xinyuan Wei and Youcheng Pan and Yang Xiang and Meishan Zhang and Haofen Wang and Jun Yu and Baotian Hu and Min Zhang},
       year={2025},
       eprint={2506.20923},
       archivePrefix={arXiv},
       primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2506.20923}, 
+      url={https://arxiv.org/abs/2506.20923},
 }
 
 @misc{hu2025kalmembedding,
-      title={KaLM-Embedding: Superior Training Data Brings A Stronger Embedding Model}, 
+      title={KaLM-Embedding: Superior Training Data Brings A Stronger Embedding Model},
       author={Xinshuo Hu and Zifei Shan and Xinping Zhao and Zetian Sun and Zhenyu Liu and Dongfang Li and Shaolin Ye and Xinyuan Wei and Qian Chen and Baotian Hu and Haofen Wang and Jun Yu and Min Zhang},
       year={2025},
       eprint={2501.01028},
       archivePrefix={arXiv},
       primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2501.01028}, 
+      url={https://arxiv.org/abs/2501.01028},
 }""",
 )
 
@@ -942,6 +953,7 @@ KaLM_Embedding_gemma_3_12b_2511 = ModelMeta(
     open_weights=True,
     release_date="2025-11-06",
     n_parameters=11.76 * 1e9,
+    n_embedding_parameters=None,
     memory_usage_mb=44884,
     max_tokens=32768,
     embed_dim=3840,
@@ -954,22 +966,22 @@ KaLM_Embedding_gemma_3_12b_2511 = ModelMeta(
     public_training_data=None,
     training_datasets=KaLM_Embedding_gemma_3_12b_training_data,
     citation="""@misc{zhao2025kalmembeddingv2,
-      title={KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model}, 
+      title={KaLM-Embedding-V2: Superior Training Techniques and Data Inspire A Versatile Embedding Model},
       author={Xinping Zhao and Xinshuo Hu and Zifei Shan and Shouzheng Huang and Yao Zhou and Xin Zhang and Zetian Sun and Zhenyu Liu and Dongfang Li and Xinyuan Wei and Youcheng Pan and Yang Xiang and Meishan Zhang and Haofen Wang and Jun Yu and Baotian Hu and Min Zhang},
       year={2025},
       eprint={2506.20923},
       archivePrefix={arXiv},
       primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2506.20923}, 
+      url={https://arxiv.org/abs/2506.20923},
 }
 
 @misc{hu2025kalmembedding,
-      title={KaLM-Embedding: Superior Training Data Brings A Stronger Embedding Model}, 
+      title={KaLM-Embedding: Superior Training Data Brings A Stronger Embedding Model},
       author={Xinshuo Hu and Zifei Shan and Xinping Zhao and Zetian Sun and Zhenyu Liu and Dongfang Li and Shaolin Ye and Xinyuan Wei and Qian Chen and Baotian Hu and Haofen Wang and Jun Yu and Min Zhang},
       year={2025},
       eprint={2501.01028},
       archivePrefix={arXiv},
       primaryClass={cs.CL},
-      url={https://arxiv.org/abs/2501.01028}, 
+      url={https://arxiv.org/abs/2501.01028},
 }""",
 )

mteb/models/model_implementations/kblab.py

@@ -10,6 +10,7 @@ sbert_swedish = ModelMeta(
     revision="6b5e83cd29c03729cfdc33d13b1423399b0efb5c",
     release_date="2023-01-11",
     n_parameters=124690944,
+    n_embedding_parameters=38_649_600,
     memory_usage_mb=476,
     embed_dim=768,
     license="apache-2.0",

mteb/models/model_implementations/kennethenevoldsen_models.py

@@ -12,6 +12,7 @@ dfm_enc_large = ModelMeta(
     revision="132c53391e7a780dc6a2f9a03724d0158fe7122c",
     release_date="2023-07-12",
     n_parameters=355087360,
+    n_embedding_parameters=51_200_000,
     memory_usage_mb=1554,
     embed_dim=1024,
     license="mit",
@@ -47,6 +48,7 @@ dfm_enc_med = ModelMeta(
     revision="701bce95d499fa97610d57e8823c54fd1fb79930",
     release_date="2023-07-12",
     n_parameters=124445952,
+    n_embedding_parameters=38_403_840,
     memory_usage_mb=475,
     embed_dim=768,
     license="mit",

mteb/models/model_implementations/kfst.py

@@ -10,6 +10,7 @@ xlmr_scandi = ModelMeta(
     revision="d40c10ca7b1e68b5a8372f2d112dac9eb3279df1",
     release_date="2022-02-22",
     n_parameters=278043648,
+    n_embedding_parameters=192_001_536,
     memory_usage_mb=1061,
     embed_dim=768,
     license="not specified",