mteb 2.1.10__py3-none-any.whl → 2.1.12__py3-none-any.whl

mteb/benchmarks/benchmarks/benchmarks.py

@@ -1647,7 +1647,7 @@ MTEB_NL = Benchmark(
             exclusive_language_filter=True,
             tasks=[
                 # Classification
-                "DutchBookReviewSentimentClassification",
+                "DutchBookReviewSentimentClassification.v2",
                 "MassiveIntentClassification",
                 "MassiveScenarioClassification",
                 "SIB200Classification",
@@ -1678,10 +1678,10 @@ MTEB_NL = Benchmark(
                 # # Reranking
                 "WikipediaRerankingMultilingual",
                 # # Retrieval
-                "ArguAna-NL",
-                "SCIDOCS-NL",
-                "SciFact-NL",
-                "NFCorpus-NL",
+                "ArguAna-NL.v2",
+                "SCIDOCS-NL.v2",
+                "SciFact-NL.v2",
+                "NFCorpus-NL.v2",
                 "BelebeleRetrieval",
                 "WebFAQRetrieval",
                 "DutchNewsArticlesRetrieval",

mteb/models/model_implementations/emillykkejensen_models.py

@@ -19,7 +19,7 @@ embedding_gemma_300m_scandi = ModelMeta(
     public_training_data="https://huggingface.co/datasets/DDSC/nordic-embedding-training-data",
     training_datasets=set(),
     similarity_fn_name="cosine",  # type: ignore[arg-type]
-    adapted_from="emillykkejensen/EmbeddingGemma-Scandi-300m",
+    adapted_from="google/embeddinggemma-300m",
     memory_usage_mb=578,
 )
 

mteb/models/model_implementations/jasper_models.py

@@ -7,14 +7,34 @@ 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 (
+    bge_chinese_training_data,
+    bge_full_data,
+    bge_m3_training_data,
+)
+from mteb.models.model_implementations.e5_instruct import E5_MISTRAL_TRAINING_DATA
+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 .nvidia_models import nvidia_training_datasets
-
 logger = logging.getLogger(__name__)
 
 
+def instruction_template(
+    instruction: str, prompt_type: PromptType | None = None
+) -> str:
+    if not instruction or prompt_type == PromptType.document:
+        return ""
+    if isinstance(instruction, dict):
+        if prompt_type is None:
+            instruction = "Given a web search query, retrieve relevant passages that answer the query"
+        else:
+            instruction = instruction[prompt_type]
+    return f"Instruct: {instruction}\nQuery:"
+
+
 class JasperModel(AbsEncoder):
     def __init__(
         self,
@@ -114,3 +134,34 @@ jasper_en_v1 = ModelMeta(
 }
 """,
 )
+
+Jasper_Token_Compression_600M = ModelMeta(
+    loader=InstructSentenceTransformerModel,
+    loader_kwargs=dict(
+        instruction_template=instruction_template,
+        apply_instruction_to_passages=False,
+        trust_remote_code=True,
+    ),
+    name="infgrad/Jasper-Token-Compression-600M",
+    languages=["eng-Latn", "zho-Hans"],
+    open_weights=True,
+    revision="06a100f753a5a96d9e583b3af79c6fcdfacc4719",
+    release_date="2025-11-14",
+    n_parameters=595776512,
+    memory_usage_mb=2272,
+    embed_dim=2048,
+    license="mit",
+    max_tokens=32768,
+    reference="https://huggingface.co/infgrad/Jasper-Token-Compression-600M",
+    similarity_fn_name="cosine",
+    framework=["Sentence Transformers", "PyTorch"],
+    use_instructions=True,
+    public_training_code=None,
+    # public_training_data: unsupervised data for distillation
+    public_training_data="https://huggingface.co/datasets/infgrad/jasper_text_distill_dataset",
+    training_datasets=bge_m3_training_data
+    | bge_chinese_training_data
+    | bge_full_data
+    | E5_MISTRAL_TRAINING_DATA
+    | qzhou_training_data,
+)

{mteb-2.1.10.dist-info → mteb-2.1.12.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mteb
-Version: 2.1.10
+Version: 2.1.12
 Summary: Massive Text Embedding Benchmark
 Author-email: MTEB Contributors <niklas@huggingface.co>, Kenneth Enevoldsen <kenneth.enevoldsen@cas.au.dk>, Nouamane Tazi <nouamane@huggingface.co>, Nils Reimers <info@nils-reimers.de>
 Maintainer-email: Kenneth Enevoldsen <kenneth.enevoldsen@cas.au.dk>, Roman Solomatin <risolomatin@gmail.com>, Isaac Chung <chungisaac1217@gmail.com>

{mteb-2.1.10.dist-info → mteb-2.1.12.dist-info}/RECORD

@@ -56,7 +56,7 @@ mteb/benchmarks/_create_table.py,sha256=z3iqa5dajLk0DYxEE9EeO1qpR3VJXokg8ZQ2rdUk
 mteb/benchmarks/benchmark.py,sha256=70RlMyyg_wkWTlU_IbfLl-KaqRWXGCKTd8fWe9X-AQE,4173
 mteb/benchmarks/get_benchmark.py,sha256=-n_O-gitRKZi48gJKNgGuI36hsP7yLVSiwulnMHN7Gw,3935
 mteb/benchmarks/benchmarks/__init__.py,sha256=UD6YjWPDVPSQdUhmD-4rho08Gs5LU9pS_C2jX5eUns0,2102
-mteb/benchmarks/benchmarks/benchmarks.py,sha256=v7n2fPGOC66zzBhS1nfSthY55DQnGNg-hGa4XfT21Vg,89941
+mteb/benchmarks/benchmarks/benchmarks.py,sha256=KDJanVYs3BkFn74VHwarZ8HJ2DX6EIgcVYBrlyjbv9I,89956
 mteb/benchmarks/benchmarks/rteb_benchmarks.py,sha256=QnCSrTTaBfcRlAQp2Nu81tgv1idMXqiM16Fp2zKJ5Ys,10607
 mteb/cli/__init__.py,sha256=v-csUr3eUZElIvrGB6QGtaIdndDfNWEe9oZchsGsJpg,64
 mteb/cli/_display_tasks.py,sha256=7A06dT9sSoTz6shyMvskPxuc5eHY_H7PGPlROzMP0yw,2196
@@ -1469,7 +1469,7 @@ mteb/models/model_implementations/dino_models.py,sha256=QFgaFHR5YKrylqJGSljXCBn2
 mteb/models/model_implementations/e5_instruct.py,sha256=9R4GoSFicgqNDCh3HhTN_8L1qhzuEKvatjHYn3T9zlU,7676
 mteb/models/model_implementations/e5_models.py,sha256=vsqkmm6XzZn9ROj_OUR0j2KiN75MEuQsOPeoyc1AeYg,10937
 mteb/models/model_implementations/e5_v.py,sha256=_9W7I0ryIzx_H9eCkzwdm8iHdGX1LIjKGXkhSh_zNv8,6690
-mteb/models/model_implementations/emillykkejensen_models.py,sha256=1DEAGdSZZXDFbbw0YH-vkLm9Y-wthgbOJCRTIpV3Jeo,2795
+mteb/models/model_implementations/emillykkejensen_models.py,sha256=QdhGqCm_1-AURkrniZj2S1MjwwIVOPMzLvpgfJq-3EQ,2779
 mteb/models/model_implementations/en_code_retriever.py,sha256=leZ-0M6LrunocY3XQBYZU1uevDRopeyR5ujIhwqBbd8,1043
 mteb/models/model_implementations/evaclip_models.py,sha256=cPMGYLDIq4s8zJxb4vPXqJ-rqwPaq7KOh2QZSO6cDas,8000
 mteb/models/model_implementations/fa_models.py,sha256=WGal70_ezITWoNdjcMdbOCTSCtoaXzuPadYstLVXxhg,7478
@@ -1483,7 +1483,7 @@ mteb/models/model_implementations/hinvec_models.py,sha256=I_d_dSNVaGIwMIwyvTlaPA
 mteb/models/model_implementations/human.py,sha256=klMpuMAtYH92EIEwNMEhne_Baf9fNiTg1DNWYD11P44,532
 mteb/models/model_implementations/ibm_granite_models.py,sha256=YCT0jbgawy19ps5l8QlxpQoJLjq8Nh-3R-e6yxS0DRM,7902
 mteb/models/model_implementations/inf_models.py,sha256=lvXUFhAYDltq2_Xa9MHcwfhh1V20rbJLSgON76tkj6w,2906
-mteb/models/model_implementations/jasper_models.py,sha256=KzjVnQ1HwaVO9Z7kk1ZkjFrhvlKupeWCmkSljnZv-IM,4071
+mteb/models/model_implementations/jasper_models.py,sha256=yf6gNPTWl05rAJrao8lIpw0wld6xdmPx9PhDwbGHSlc,6037
 mteb/models/model_implementations/jina_clip.py,sha256=CfiIxbhKspjQajNtObCfGPHOWPk6uLn4cuwydQHFTMo,5118
 mteb/models/model_implementations/jina_models.py,sha256=QWoesiTygdFTLcdGpdx26wOUI1AXRz3jLmxGHJ0WMNE,29919
 mteb/models/model_implementations/kalm_models.py,sha256=FmW7Z5Qs6WYBLuKvql3u4IJW36kj4k-Ypah8qTBEBkg,59837
@@ -2554,9 +2554,9 @@ mteb/types/_metadata.py,sha256=NN-W0S6a5TDV7UkpRx1pyWtGF4TyyCyoPUfHOwdeci8,2290
 mteb/types/_result.py,sha256=CRAUc5IvqI3_9SyXDwv-PWLCXwXdZem9RePeYESRtuw,996
 mteb/types/_string_validators.py,sha256=PY-dYq4E8O50VS3bLYdldPWp400fl_WzUjfVSkNWe8U,523
 mteb/types/statistics.py,sha256=YwJsxTf1eaCI_RE-J37a-gK5wDeGAsmkeZKoZCFihSo,3755
-mteb-2.1.10.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-mteb-2.1.10.dist-info/METADATA,sha256=LClBepxtjXoGssnPn6QgdAukEqJerTX67OC7zoKhdiE,13574
-mteb-2.1.10.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mteb-2.1.10.dist-info/entry_points.txt,sha256=8IJoEJFKoDHmVnNev-qJ9pp4Ln7_1-ma9QsXnzVCzGU,39
-mteb-2.1.10.dist-info/top_level.txt,sha256=OLVIjcQAlWBz0bdmutKlWHLF42FF0hp4uVAg3ZyiG4U,5
-mteb-2.1.10.dist-info/RECORD,,
+mteb-2.1.12.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+mteb-2.1.12.dist-info/METADATA,sha256=mXrsN01rI1osGl_9epUwEI7BjLmwXSxJECQjR7BmoJM,13574
+mteb-2.1.12.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mteb-2.1.12.dist-info/entry_points.txt,sha256=8IJoEJFKoDHmVnNev-qJ9pp4Ln7_1-ma9QsXnzVCzGU,39
+mteb-2.1.12.dist-info/top_level.txt,sha256=OLVIjcQAlWBz0bdmutKlWHLF42FF0hp4uVAg3ZyiG4U,5
+mteb-2.1.12.dist-info/RECORD,,