mteb 2.6.9__py3-none-any.whl → 2.7.1__py3-none-any.whl

mteb/tasks/retrieval/multilingual/vidore3_bench_retrieval.py

@@ -15,7 +15,7 @@ class Vidore3FinanceEnRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3FinanceEnRetrieval",
         description="Retrieve associated pages according to questions. This task, Finance - EN, is a corpus of reports from american banking companies, intended for long-document understanding tasks. Original queries were created in english, then translated to french, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_finance_en_mteb_format",
             "revision": "fa78cb14152b3dde8c5defdc4e3ddf50de69dfeb",
@@ -34,15 +34,14 @@ class Vidore3FinanceEnRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -53,7 +52,7 @@ class Vidore3FinanceFrRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3FinanceFrRetrieval",
         description="Retrieve associated pages according to questions. This task, Finance - FR, is a corpus of reports from french companies in the luxury domain, intended for long-document understanding tasks. Original queries were created in french, then translated to english, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_finance_fr_mteb_format",
             "revision": "8a2adfda85a7967c7252129703d9b3c7c9f038a9",
@@ -71,15 +70,14 @@ class Vidore3FinanceFrRetrieval(AbsTaskRetrieval):
         dialect=[],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -91,7 +89,7 @@ class Vidore3IndustrialRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3IndustrialRetrieval",
         description="Retrieve associated pages according to questions. This dataset, Industrial reports, is a corpus of technical documents on military aircraft (fueling, mechanics...), intended for complex-document understanding tasks. Original queries were created in english, then translated to french, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_industrial_mteb_format",
             "revision": "f732b725cf4a70803210edfe265a04f8bd5328f6",
@@ -110,15 +108,14 @@ class Vidore3IndustrialRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -130,7 +127,7 @@ class Vidore3PharmaceuticalsRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3PharmaceuticalsRetrieval",
         description="Retrieve associated pages according to questions. This dataset, Pharmaceutical, is a corpus of slides from the FDA, intended for long-document understanding tasks. Original queries were created in english, then translated to french, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_pharmaceuticals_mteb_format",
             "revision": "237ed4f43c7fb3c4df07ec4e9dd0a4366be555b0",
@@ -149,15 +146,14 @@ class Vidore3PharmaceuticalsRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -169,7 +165,7 @@ class Vidore3ComputerScienceRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3ComputerScienceRetrieval",
         description="Retrieve associated pages according to questions. This dataset, Computer Science, is a corpus of textbooks from the openstacks website, intended for long-document understanding tasks. Original queries were created in english, then translated to french, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_computer_science_mteb_format",
             "revision": "fb7fb69f81f7db62790f40494124b8ad22b424ab",
@@ -188,15 +184,14 @@ class Vidore3ComputerScienceRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -208,7 +203,7 @@ class Vidore3HrRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3HrRetrieval",
         description="Retrieve associated pages according to questions. This dataset, HR, is a corpus of reports released by the european union, intended for complex-document understanding tasks. Original queries were created in english, then translated to french, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_hr_mteb_format",
             "revision": "bc7d43d64815ed30f664168c8052106484aba7fd",
@@ -227,15 +222,14 @@ class Vidore3HrRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -247,7 +241,7 @@ class Vidore3EnergyRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3EnergyRetrieval",
         description="Retrieve associated pages according to questions. This dataset, Energy Fr, is a corpus of reports on energy supply in europe, intended for complex-document understanding tasks. Original queries were created in french, then translated to english, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_energy_mteb_format",
             "revision": "84fca99e5978604bae30f2436eacb6dbaa0532e9",
@@ -266,15 +260,14 @@ class Vidore3EnergyRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -286,7 +279,7 @@ class Vidore3PhysicsRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3PhysicsRetrieval",
         description="Retrieve associated pages according to questions. This dataset, Physics, is a corpus of course slides on french bachelor level physics lectures, intended for complex visual understanding tasks. Original queries were created in french, then translated to english, german, italian, portuguese and spanish.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "vidore/vidore_v3_physics_mteb_format",
             "revision": "2c18ef90ab3ef93a9d86ecc6521cdae2a29f8300",
@@ -305,15 +298,14 @@ class Vidore3PhysicsRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -325,7 +317,7 @@ class Vidore3NuclearRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3NuclearRetrieval",
         description="Retrieve associated pages according to questions.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "mteb-private/Vidore3NuclearRetrieval",
             "revision": "a463fc67fefc01152153101e88a32d5f9515e3e3",
@@ -344,15 +336,14 @@ class Vidore3NuclearRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},
@@ -364,7 +355,7 @@ class Vidore3TelecomRetrieval(AbsTaskRetrieval):
     metadata = TaskMetadata(
         name="Vidore3TelecomRetrieval",
         description="Retrieve associated pages according to questions.",
-        reference="https://huggingface.co/blog/QuentinJG/introducing-vidore-v3",
+        reference="https://arxiv.org/abs/2601.08620",
         dataset={
             "path": "mteb-private/Vidore3TelecomRetrieval",
             "revision": "a54635a274ef2835721b7cbe3eb27483b9ec964b",
@@ -383,15 +374,14 @@ class Vidore3TelecomRetrieval(AbsTaskRetrieval):
         modalities=["text", "image"],
         sample_creation="created and machine-translated",
         bibtex_citation=r"""
-@misc{mace2025vidorev3,
-  author = {Macé, Quentin and Loison, Antonio and EDY, Antoine and Xing, Victor and Viaud, Gautier},
-  day = {5},
-  howpublished = {\url{https://huggingface.co/blog/QuentinJG/introducing-vidore-v3}},
-  journal = {Hugging Face Blog},
-  month = {November},
-  publisher = {Hugging Face},
-  title = {ViDoRe V3: a comprehensive evaluation of retrieval for enterprise use-cases},
-  year = {2025},
+@article{loison2026vidorev3comprehensiveevaluation,
+  archiveprefix = {arXiv},
+  author = {António Loison and Quentin Macé and Antoine Edy and Victor Xing and Tom Balough and Gabriel Moreira and Bo Liu and Manuel Faysse and Céline Hudelot and Gautier Viaud},
+  eprint = {2601.08620},
+  primaryclass = {cs.AI},
+  title = {ViDoRe V3: A Comprehensive Evaluation of Retrieval Augmented Generation in Complex Real-World Scenarios},
+  url = {https://arxiv.org/abs/2601.08620},
+  year = {2026},
 }
 """,
         prompt={"query": "Find a screenshot that is relevant to the user's question."},

{mteb-2.6.9.dist-info → mteb-2.7.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mteb
-Version: 2.6.9
+Version: 2.7.1
 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>
@@ -32,6 +32,8 @@ Requires-Dist: rich>=0.0.0
 Requires-Dist: pytrec-eval-terrier>=0.5.6
 Requires-Dist: pydantic>=2.0.0
 Requires-Dist: polars>=0.20.22
+Requires-Dist: torch; python_full_version < "3.14"
+Requires-Dist: torch>=2.9.0; python_full_version >= "3.14"
 Provides-Extra: image
 Requires-Dist: torchvision>0.2.1; extra == "image"
 Requires-Dist: transformers[torch-vision,vision]; extra == "image"
@@ -108,6 +110,8 @@ Provides-Extra: faiss-cpu
 Requires-Dist: faiss-cpu>=1.12.0; extra == "faiss-cpu"
 Provides-Extra: eager-embed
 Requires-Dist: qwen_vl_utils>=0.0.14; extra == "eager-embed"
+Provides-Extra: vllm
+Requires-Dist: vllm>=0.11.1; extra == "vllm"
 Dynamic: license-file
 
 <h1 align="center">

{mteb-2.6.9.dist-info → mteb-2.7.1.dist-info}/RECORD

@@ -21,7 +21,7 @@ mteb/_evaluators/clustering_evaluator.py,sha256=YfjwpZL03Tiu0jfd6HBSavrOs_WALwIp
 mteb/_evaluators/evaluator.py,sha256=CZIFM84KJcbygBqHrd6Reg0FA1ZT5dfVjFTuBYgGyMw,906
 mteb/_evaluators/pair_classification_evaluator.py,sha256=kgxvnzsZPFF7c6GNRWtyAP3AJ3v_heRQUYwmYvNATzI,6440
 mteb/_evaluators/retrieval_evaluator.py,sha256=UVrGtjOQdbDD2ZeH1oVm9ouIlU7Uup2a09ylzm0-QaM,3024
-mteb/_evaluators/retrieval_metrics.py,sha256=jd5BEinfGPdHPK3kXf6kfBs0gvDuBXEWyx-RS7G4c8g,23756
+mteb/_evaluators/retrieval_metrics.py,sha256=fwsPemssWqQRvo0t4PZTGrrv5KXK8N6U35kOFSkRbEU,23755
 mteb/_evaluators/sklearn_evaluator.py,sha256=lipgxkGXWeKGD2jhaNL9WQqLBS1_2L_WZpUobZR7iBI,3838
 mteb/_evaluators/zeroshot_classification_evaluator.py,sha256=Y5ZFUsBAWXMpKnZ6Iw1K0lm0_L4chLMxC4LAdZCBbPs,2296
 mteb/_evaluators/image/__init__.py,sha256=CsQd7OMkeV2Phun7paPWjayZ5qRnvj8H0TYBFeqMxag,148
@@ -43,7 +43,7 @@ mteb/abstasks/multilabel_classification.py,sha256=D_bdK3yyZFgaH1pg9PpO5LHQMVaJP3
 mteb/abstasks/pair_classification.py,sha256=-T8q7JQRjHZ3ihLqnuR2yuuci9xljvUqCldsHGnGLGc,13767
 mteb/abstasks/regression.py,sha256=sROjvfasLS89KRPUTsc1ONFsBTzfoqlpxLQfIkKBQXs,8763
 mteb/abstasks/retrieval.py,sha256=keuf7GeYyM0ODPlVFGZIFBmBoSil0c1glmcu6C0YzhA,26921
-mteb/abstasks/retrieval_dataset_loaders.py,sha256=WukcFAn54rUpXULCG43eysHozXHAxo2CaPhQyL_2Yg8,9401
+mteb/abstasks/retrieval_dataset_loaders.py,sha256=Jcr2hHe5sqWc8eL3WDhR9J7bJokD7tyhz_z3eD14o0M,9403
 mteb/abstasks/sts.py,sha256=b_19QvclU0_q0aEJdfvB3weUQ96bUS7uDuuRrjpz8Bc,9245
 mteb/abstasks/task_metadata.py,sha256=7TM_ls5bzYA1dHFq3VQgeioiyLrvMQz4i3hmWIsnD4M,27029
 mteb/abstasks/zeroshot_classification.py,sha256=O8jxoBgnrRx-BzOnr1aJVK3wIEBwkA4xuMxdVxlzJqI,6053
@@ -54,14 +54,14 @@ mteb/abstasks/image/__init__.py,sha256=NgvMJnp1g2mUv27RL-TvzA7s1BOdMG-EB1CrZfdbW
 mteb/abstasks/image/image_text_pair_classification.py,sha256=4RfPdAlb4ZlAE-9DiFQUvm7YpcQg9kTY7HRjRVenLmk,7935
 mteb/abstasks/text/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 mteb/abstasks/text/bitext_mining.py,sha256=0Ff7t1jMEonajHzB5CzZZoSMVkC5KTa0muoytcNzOjM,11152
-mteb/abstasks/text/reranking.py,sha256=QMgAAndGYRzvQdlhjLRMxrh_yrJZ0VQH40I-7mXo1O0,7872
+mteb/abstasks/text/reranking.py,sha256=vD5YUwst--zJ01bU40Z7juqQyuv8jrareRTtoCwt8E4,7871
 mteb/abstasks/text/summarization.py,sha256=qCUWfxaVAW0EYTFClUh9puBMGttyV6C7qKpCcHPAZN8,7148
 mteb/benchmarks/__init__.py,sha256=MQEVeli-zLaJ7Xg0z7RhXQwsdmm7Ht_W2Ln0rZo1Szc,225
 mteb/benchmarks/_create_table.py,sha256=b2RqGqi0ZonKbHecEcZiF4pkfE96smFRIzxOI82ETA8,22304
 mteb/benchmarks/benchmark.py,sha256=RheQOo0iQbu_ylN7oFLr2r-z_ahrMCTvKscVuwUx6yo,5694
 mteb/benchmarks/get_benchmark.py,sha256=nzR6cu5yXu1kIJKhd4A2R62xp43Z62bluPbOpNXHMWQ,2545
-mteb/benchmarks/benchmarks/__init__.py,sha256=NeFt0CGdY25-UWJ2aGCjDnpzr5-geih_1jjp1GQXJz4,2256
-mteb/benchmarks/benchmarks/benchmarks.py,sha256=w62HyautMAuRrwxUHjGpDqxTrXkb2zTD5cqXJV29Fyo,99644
+mteb/benchmarks/benchmarks/__init__.py,sha256=767VzDgayFSB3KJ05o0gFdOeo2f1EarbEfnCsy_karw,2294
+mteb/benchmarks/benchmarks/benchmarks.py,sha256=zAm1NTggaWb5Am6O30FUf-8zIbbF39eiuhSyLHzlpxQ,102370
 mteb/benchmarks/benchmarks/rteb_benchmarks.py,sha256=QnCSrTTaBfcRlAQp2Nu81tgv1idMXqiM16Fp2zKJ5Ys,10607
 mteb/cli/__init__.py,sha256=v-csUr3eUZElIvrGB6QGtaIdndDfNWEe9oZchsGsJpg,64
 mteb/cli/_display_tasks.py,sha256=pWKupzak8uxEIwJZbYpZpteeVprOgVT9Wr0HYeypitQ,2206
@@ -1082,6 +1082,7 @@ mteb/descriptive_stats/Retrieval/CUREv1.json,sha256=lfz-JMLuE5KlWqM08CUF80Nex1Oh
 mteb/descriptive_stats/Retrieval/ChatDoctorRetrieval.json,sha256=VyyOcmc9GQ1zqXrvAv0pFR5OAK-pE0fQYGJOsl4Wvok,991
 mteb/descriptive_stats/Retrieval/ChemHotpotQARetrieval.json,sha256=V2oT8C6Wh5VR6LeBlxrYH0VUJeEisjJELOZHpN2NP1o,2950
 mteb/descriptive_stats/Retrieval/ChemNQRetrieval.json,sha256=6E8SefjKJdqVC1R0JvGOPrJdpogDT-1H9X19GryVfI8,985
+mteb/descriptive_stats/Retrieval/ChemRxivRetrieval.json,sha256=kIfyVNgoLtO1QpnKguIovQVUio8RVdc1bBY_kwogT-Q,986
 mteb/descriptive_stats/Retrieval/ClimateFEVER-Fa.json,sha256=B1bp66dEPtY6p1zgHTK3fyBN_eFbmprpjhwUovVs4mI,1016
 mteb/descriptive_stats/Retrieval/ClimateFEVER-NL.json,sha256=RaHI8NhlXCD28FGFd1-Pofm4lvD-_Kcszde2OgPxQk8,1017
 mteb/descriptive_stats/Retrieval/ClimateFEVER-VN.json,sha256=D2xgItr4sBQiW3x8qr12firW0cXsOHygKZcDrtjhQ10,1016
@@ -1140,6 +1141,7 @@ mteb/descriptive_stats/Retrieval/EnglishFinance3Retrieval.json,sha256=IpCB7e6cwO
 mteb/descriptive_stats/Retrieval/EnglishFinance4Retrieval.json,sha256=tdSyer5_q9rPZSD0LEjjXluCbWY_W_FewhOTQycU1U0,995
 mteb/descriptive_stats/Retrieval/EnglishHealthcare1Retrieval.json,sha256=c_QduudRLaQVFARJxSc-YTFwHeNyhiu7IMv6JGlYC5c,1005
 mteb/descriptive_stats/Retrieval/EstQA.json,sha256=qqmmX7ExWg4152S6yiUAyHzaLS0I-QUaOn9KVBZ3UZ8,981
+mteb/descriptive_stats/Retrieval/EuroPIRQRetrieval.json,sha256=FIpF0h7hhF7O4jn-rEa9a-_8tPD1MAY-klTkSX3_SI8,4586
 mteb/descriptive_stats/Retrieval/FEVER-FaHardNegatives.json,sha256=KMDX6lg671e76rAboz30KgpvJAPxVSa6Thgefj1tgM4,987
 mteb/descriptive_stats/Retrieval/FEVER-NL.json,sha256=K810hO-zNJWg3-i30oma72wsjORGu8lLLd4PN_w0VVs,1020
 mteb/descriptive_stats/Retrieval/FEVER-VN.json,sha256=MnadQ0FGkXCFCDCI8eGwFYbouo8ta_1W_WlrqNx1clI,1018
@@ -1459,10 +1461,11 @@ mteb/models/__init__.py,sha256=ABTuoqiBjBtBWW3LYY7ItBHdylR6jWoy06HH0g6j6fU,910
 mteb/models/abs_encoder.py,sha256=6e9UAk7ckYsJ6hItIPMfSaPdlX_FxfJ-OdJ87oqT2OM,16625
 mteb/models/get_model_meta.py,sha256=wVh2FaWevJ10hJlbm-FQtTQazLMfnkEV3IK7PUyBPOQ,6082
 mteb/models/instruct_wrapper.py,sha256=GLHg9KcgYu2rF15LEMKesRpPudGfKE2y-aLXVG_CLj0,9670
-mteb/models/model_meta.py,sha256=x8EuA8Zpc4DqhK_50v5TAZ7n2J2yhHqf5U0ldCpPnw0,31101
+mteb/models/model_meta.py,sha256=BDT4Q_aYPm3RUgtXq2poe1VyUUHob_fLrDKA4BPgYrg,31143
 mteb/models/models_protocols.py,sha256=5WYOZw3-T-wK7ux0YZVCfbcMTkAisqAqbu44ZNoir4A,9250
 mteb/models/search_wrappers.py,sha256=9lXLXUyL6atMCwXp-HBUong6msT3UAUY9QI7zKXbSVU,20945
-mteb/models/sentence_transformer_wrapper.py,sha256=3sAev15a07ULA_ikFBBsMta0toy9AGPBbBPi37j_K-A,12932
+mteb/models/sentence_transformer_wrapper.py,sha256=liMzSqr0bC-yI_oEL50ckEc4ti9nibDLIiukP2crhOA,13543
+mteb/models/vllm_wrapper.py,sha256=ebX4JIKPoxW4PIlr3BnaoLGuMGRpHzBe_ZwvFscx1D0,12286
 mteb/models/cache_wrappers/__init__.py,sha256=1w1TnMwulWJSzNkLXjbh5MY3sqgHWc6vUntYn49i9X8,169
 mteb/models/cache_wrappers/cache_backend_protocol.py,sha256=iGWdqDEoaCxUVEnwsXhy-m9d2QX8KTaQ9m2ZyawrMes,1634
 mteb/models/cache_wrappers/cache_wrapper.py,sha256=GPC0UhHfkUH-i-Q4HdFvBev6h6GtMlWEId_B3tL-J54,6600
@@ -1544,8 +1547,8 @@ mteb/models/model_implementations/nbailab.py,sha256=LM00HJIr4yrA45qh2O21BIDXku9K
 mteb/models/model_implementations/no_instruct_sentence_models.py,sha256=qLiMok_OxKvIYXWnP0KNWqH1monZx-OdSZrSx3QEhtI,4049
 mteb/models/model_implementations/nomic_models.py,sha256=dmQC_cWg6hAmiBHK7fXoXEiGBJnJvrq0RsnCcJ2qe1Q,15137
 mteb/models/model_implementations/nomic_models_vision.py,sha256=usCKfZCR7aEi_DnNmVAYjH-lXx_ipQkBVtUAmhJ90QI,6870
-mteb/models/model_implementations/nvidia_llama_nemoretriever_colemb.py,sha256=6dTGtK1GiaYdpJ4IQFgCCOkGyHQyuEUatKs-Uv-1YmE,6450
-mteb/models/model_implementations/nvidia_models.py,sha256=XnNiyByz6EFrISz1Msb-cXLVQfKnYP5HMRzAXC1KPDo,24937
+mteb/models/model_implementations/nvidia_llama_nemoretriever_colemb.py,sha256=pJqkKBNns6jUYlpI3aGtpmrhXOSYgEvCje1ir_yXdpc,6889
+mteb/models/model_implementations/nvidia_models.py,sha256=KcnH7wGhz4LQ2F_y9Q9cJTr89DNKeHjVkVgHUfftzIY,26685
 mteb/models/model_implementations/octen_models.py,sha256=FFR1-yG2euN-6kgL4qJNHYB6cPsds4NGYFPmc5tHhoE,8514
 mteb/models/model_implementations/openai_models.py,sha256=905BajYi_XyOZgqU3AeKpwIttLoUitaAyc48sTWI6Jg,9482
 mteb/models/model_implementations/openclip_models.py,sha256=MyosgeYSrgBXGuGFtI2Tyxksxpb7bADFJVSYFCLweVA,11622
@@ -1591,8 +1594,8 @@ mteb/models/model_implementations/vdr_models.py,sha256=8jlfABvO7Z9ebzAPFHqln3B2I
 mteb/models/model_implementations/vi_vn_models.py,sha256=Ep2zj4Xvjyu0a_YiLsYvolKdMGSOtzm-N-yNyXmfNwA,6328
 mteb/models/model_implementations/vista_models.py,sha256=GkQFHIwwjxwM0wDuo-dWJBo4dLExlHtHfXwhcdKA5uQ,10884
 mteb/models/model_implementations/vlm2vec_models.py,sha256=EeWl3kpS_1VDJs4t1QmpaWSuglLPL2GyZu27fVY1VT8,11802
-mteb/models/model_implementations/voyage_models.py,sha256=5A5RD2A6B20qLDVEpWL0TNMQOf5hnTVXdBugdh5q4d0,20214
-mteb/models/model_implementations/voyage_v.py,sha256=eFdSOKka5VoLjViZk5umlgTw_ETjyXv4yhZ9SoCR-p0,8124
+mteb/models/model_implementations/voyage_models.py,sha256=jwVjgx9E6-rZxd8wmuNbbQxKTobuWBNWrs0ezqp4Oik,22525
+mteb/models/model_implementations/voyage_v.py,sha256=JqtXnICeaODRZHBj_Xsaf3PfIG-XPKopblNxAXHqYNo,8159
 mteb/models/model_implementations/xyz_models.py,sha256=gjwCx3U4AxMcJDTSWVoYV6xeyXLw7lUZI5D6Q7JjWho,1322
 mteb/models/model_implementations/youtu_models.py,sha256=THwWRabutW-qC-JZOVhxXWjKHVyMElzt_xm81ixzN50,5995
 mteb/models/model_implementations/yuan_models.py,sha256=j-QIKECPg4TiBW_3Bp6g5yr2UOdFziFSeoGE4uKepSM,980
@@ -2207,7 +2210,7 @@ mteb/tasks/retrieval/deu/german_qu_ad_retrieval.py,sha256=RYZCPgKaNPi14ncxnvz8_7
 mteb/tasks/retrieval/deu/legal_qu_ad_retrieval.py,sha256=l_UNVkOQxqlKp6wef2BM-GKtyYnmYLYThGZZnUsR_-c,1594
 mteb/tasks/retrieval/ell/__init__.py,sha256=46naXAZtJzyezyqOQGRIlr4zQVkqGQJdj7ztjMrez9Y,72
 mteb/tasks/retrieval/ell/greek_civics_qa.py,sha256=uKmiWHHZvL1o5m17XCrzi16kHWFNhlEUHrtlk-LR1as,2514
-mteb/tasks/retrieval/eng/__init__.py,sha256=loq6PjQTX-ZpCUgy6wFZ74LCpZVmzbiRQKtJQlJbvQM,16093
+mteb/tasks/retrieval/eng/__init__.py,sha256=dVW7pHRu8SXp2_PJwn6InbfWVQ2VVy0PJvr_-bLBrc4,16158
 mteb/tasks/retrieval/eng/aila_casedocs_retrieval.py,sha256=UKoN9oE8C412REf8MV16aUDgE5NwkHxnXsh4dcLztpk,1398
 mteb/tasks/retrieval/eng/aila_statutes_retrieval.py,sha256=GugjZwaWmTlNyYzK8ACKZHIiUw2YBvLaVyTngN_qQyM,1366
 mteb/tasks/retrieval/eng/alpha_nli_retrieval.py,sha256=GpOkizyeOs_ZMbRu9g1yAXdKkQr8PObUepP1OulbJio,1714
@@ -2228,6 +2231,7 @@ mteb/tasks/retrieval/eng/built_bench_retrieval.py,sha256=pqsIaVMl0ugGl5wx1oHPooq
 mteb/tasks/retrieval/eng/chat_doctor_retrieval.py,sha256=_PQYn3jXgeLKjnb_uKxzHtM9SCICV3bsmgEcEoD2OXo,3582
 mteb/tasks/retrieval/eng/chem_hotpot_qa_retrieval.py,sha256=sLTfqBf967htSU7Ego7zkEC8QYVWFI12YoHxa-urWEw,2114
 mteb/tasks/retrieval/eng/chem_nq_retrieval.py,sha256=8bl4PRKJwYgNF0sZPZQINgn81-r3c_2gDoMQJYdpb8I,1886
+mteb/tasks/retrieval/eng/chemrxiv.py,sha256=-HS_axsMPaEKr8T0d9WvgfJ_UmAr00InHFjFpt4VDVo,1404
 mteb/tasks/retrieval/eng/cirr_it2i_retrieval.py,sha256=o_4fVGosZjYsfMQy7tzQGaq8aijhUwXKcV7MYIYY4SY,1583
 mteb/tasks/retrieval/eng/climate_fever_retrieval.py,sha256=obIROibY5nQ8oDGbkjG5Z3jrJ-rAXVmF_1w_XN6Nqm0,4615
 mteb/tasks/retrieval/eng/cqa_dupstack_android_retrieval.py,sha256=-Uj8BOIPyjl8egm34qZGRKULGgEoaoac4wOhdWXAPgE,1674
@@ -2392,11 +2396,12 @@ mteb/tasks/retrieval/kor/auto_rag_retrieval.py,sha256=tgffW8zMpDSv1FCOdS4_4SL5zK
 mteb/tasks/retrieval/kor/ko_strategy_qa.py,sha256=jk13ORetYtF0q36h8ljD6TeTHUwvK5F5ZbDoMCP3eWk,1156
 mteb/tasks/retrieval/kor/kovidore2_bench_retrieval.py,sha256=AjOLe2l9drBWOCeGzQqxfee3gwwU6ElAJ7-5pbFr6C8,6208
 mteb/tasks/retrieval/kor/squad_kor_v1_retrieval.py,sha256=M7T5FkN1efK7euRslx-LZN7hS_QdIwqtUuVlWO-dico,1631
-mteb/tasks/retrieval/multilingual/__init__.py,sha256=mfVGkoB4DO5ktlg8ia-4nImFVmZcqXh1XkgCkIff0tY,6765
+mteb/tasks/retrieval/multilingual/__init__.py,sha256=rbeuLmNYooHPjgROuEOH84Q6QmGhuXnedej0d6xAgqc,6841
 mteb/tasks/retrieval/multilingual/belebele_retrieval.py,sha256=gaVLEwuLEwMutMi9V-obpiYKbpllX2QNm2j3MVeebfE,7027
 mteb/tasks/retrieval/multilingual/cross_lingual_semantic_discrimination_wmt19.py,sha256=_6r34ZvRiLVENYcrd87NjilybGaetBwKFEbO29zYmBU,4676
 mteb/tasks/retrieval/multilingual/cross_lingual_semantic_discrimination_wmt21.py,sha256=Puy0PjpRr4M_Bbxdl7oWfa7pQGM04zaRaTNlnhyKejM,4677
 mteb/tasks/retrieval/multilingual/cur_ev1_retrieval.py,sha256=dwzo2sqjamM_xkSiC-jbapyhDFezSJpM4S8KfBsuLPk,4562
+mteb/tasks/retrieval/multilingual/euro_pirq_retrieval.py,sha256=rlbgWMRdQgDD8z4ZBPgGU1fRAqjmoFDzh0uD_P6qR-4,1602
 mteb/tasks/retrieval/multilingual/indic_qa_retrieval.py,sha256=K7iWZ-yTftZFQiXBOlkTJXGpQXs-ZFt6OQj_L6HjEwk,1872
 mteb/tasks/retrieval/multilingual/jina_vdr_bench_retrieval.py,sha256=dw2YuFrA5OWFTL5zREiix93oAj7WcpcnAhCRh8YRoHI,44579
 mteb/tasks/retrieval/multilingual/mintaka_retrieval.py,sha256=SwOliONITZM679LIBSMrvx_VymqE-zRN6YiYahhzfzw,2229
@@ -2413,7 +2418,7 @@ mteb/tasks/retrieval/multilingual/ru_sci_bench_retrieval.py,sha256=Mmcvrt_1cIxPf
 mteb/tasks/retrieval/multilingual/statcan_dialogue_dataset_retrieval.py,sha256=iFUQUlO_ogBdQBVYBQW3o-AJDQ792yg1pJtRxA5I3Qo,3796
 mteb/tasks/retrieval/multilingual/vdr_multilingual_retrieval.py,sha256=UduWKefwP7bPYxiDlztPEvSWXmTdw0xElglMbPY6XhA,4449
 mteb/tasks/retrieval/multilingual/vidore2_bench_retrieval.py,sha256=vOfiruHywYkP8pccdAuGLyYyFTw1zK0qcXDnUFA8Z5A,9091
-mteb/tasks/retrieval/multilingual/vidore3_bench_retrieval.py,sha256=wOoC--IVTz0dR6RMVICbz6OWxfCyVahGDSfX_TScCgA,16934
+mteb/tasks/retrieval/multilingual/vidore3_bench_retrieval.py,sha256=V3jtSlWhoKR1PCvHsH0HrONy-oFghomwqihBonQs_50,17414
 mteb/tasks/retrieval/multilingual/web_faq_retrieval.py,sha256=TM-Q98yXZny_PKHAFNEvw9o9ET_L6VM3aNis1NJ9DgM,2686
 mteb/tasks/retrieval/multilingual/wikipedia_retrieval_multilingual.py,sha256=zyqAt63bHXNU_I37jb891pwWUyGzZUGkXCyhWlRbed8,1569
 mteb/tasks/retrieval/multilingual/wit_t2i_retrieval.py,sha256=_swZhhMRs5OhzBdJVqQF1i9ZrTvAxaVrG0TpkPWkoHo,4359
@@ -2617,9 +2622,9 @@ mteb/types/_metadata.py,sha256=NN-W0S6a5TDV7UkpRx1pyWtGF4TyyCyoPUfHOwdeci8,2290
 mteb/types/_result.py,sha256=UKNokV9pu3G74MGebocU512aU_fFU9I9nPKnrG9Q0iE,1035
 mteb/types/_string_validators.py,sha256=PY-dYq4E8O50VS3bLYdldPWp400fl_WzUjfVSkNWe8U,523
 mteb/types/statistics.py,sha256=GwkBPmAr18Onu-vHtzHs0PFrhCozdOMiT13HwnWL4ZM,3961
-mteb-2.6.9.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-mteb-2.6.9.dist-info/METADATA,sha256=l-i7rdEestHq4eEfqtwTtdvHS7le1gDwi4EMzu0NW4g,14281
-mteb-2.6.9.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mteb-2.6.9.dist-info/entry_points.txt,sha256=8IJoEJFKoDHmVnNev-qJ9pp4Ln7_1-ma9QsXnzVCzGU,39
-mteb-2.6.9.dist-info/top_level.txt,sha256=OLVIjcQAlWBz0bdmutKlWHLF42FF0hp4uVAg3ZyiG4U,5
-mteb-2.6.9.dist-info/RECORD,,
+mteb-2.7.1.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+mteb-2.7.1.dist-info/METADATA,sha256=olR8WXq0vPmp5FUUNww2hAMIuhr5irC4wnymgFN0xhY,14457
+mteb-2.7.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mteb-2.7.1.dist-info/entry_points.txt,sha256=8IJoEJFKoDHmVnNev-qJ9pp4Ln7_1-ma9QsXnzVCzGU,39
+mteb-2.7.1.dist-info/top_level.txt,sha256=OLVIjcQAlWBz0bdmutKlWHLF42FF0hp4uVAg3ZyiG4U,5
+mteb-2.7.1.dist-info/RECORD,,