MMTEB: Massive Multilingual Text Embedding Benchmark
Paper • 2502.13595 • Published • 50
Datasets:
id stringlengths 1 3 | text stringlengths 29 68 |
|---|---|
0 | a photo of a am general hummer suv 2000. |
1 | a photo of a acura rl sedan 2012. |
2 | a photo of a acura tl sedan 2012. |
3 | a photo of a acura tl type-s 2008. |
4 | a photo of a acura tsx sedan 2012. |
5 | a photo of a acura integra type r 2001. |
6 | a photo of a acura zdx hatchback 2012. |
7 | a photo of a aston martin v8 vantage convertible 2012. |
8 | a photo of a aston martin v8 vantage coupe 2012. |
9 | a photo of a aston martin virage convertible 2012. |
10 | a photo of a aston martin virage coupe 2012. |
11 | a photo of a audi rs 4 convertible 2008. |
12 | a photo of a audi a5 coupe 2012. |
13 | a photo of a audi tts coupe 2012. |
14 | a photo of a audi r8 coupe 2012. |
15 | a photo of a audi v8 sedan 1994. |
16 | a photo of a audi 100 sedan 1994. |
17 | a photo of a audi 100 wagon 1994. |
18 | a photo of a audi tt hatchback 2011. |
19 | a photo of a audi s6 sedan 2011. |
20 | a photo of a audi s5 convertible 2012. |
21 | a photo of a audi s5 coupe 2012. |
22 | a photo of a audi s4 sedan 2012. |
23 | a photo of a audi s4 sedan 2007. |
24 | a photo of a audi tt rs coupe 2012. |
25 | a photo of a bmw activehybrid 5 sedan 2012. |
26 | a photo of a bmw 1 series convertible 2012. |
27 | a photo of a bmw 1 series coupe 2012. |
28 | a photo of a bmw 3 series sedan 2012. |
29 | a photo of a bmw 3 series wagon 2012. |
30 | a photo of a bmw 6 series convertible 2007. |
31 | a photo of a bmw x5 suv 2007. |
32 | a photo of a bmw x6 suv 2012. |
33 | a photo of a bmw m3 coupe 2012. |
34 | a photo of a bmw m5 sedan 2010. |
35 | a photo of a bmw m6 convertible 2010. |
36 | a photo of a bmw x3 suv 2012. |
37 | a photo of a bmw z4 convertible 2012. |
38 | a photo of a bentley continental supersports conv. convertible 2012. |
39 | a photo of a bentley arnage sedan 2009. |
40 | a photo of a bentley mulsanne sedan 2011. |
41 | a photo of a bentley continental gt coupe 2012. |
42 | a photo of a bentley continental gt coupe 2007. |
43 | a photo of a bentley continental flying spur sedan 2007. |
44 | a photo of a bugatti veyron 16.4 convertible 2009. |
45 | a photo of a bugatti veyron 16.4 coupe 2009. |
46 | a photo of a buick regal gs 2012. |
47 | a photo of a buick rainier suv 2007. |
48 | a photo of a buick verano sedan 2012. |
49 | a photo of a buick enclave suv 2012. |
50 | a photo of a cadillac cts-v sedan 2012. |
51 | a photo of a cadillac srx suv 2012. |
52 | a photo of a cadillac escalade ext crew cab 2007. |
53 | a photo of a chevrolet silverado 1500 hybrid crew cab 2012. |
54 | a photo of a chevrolet corvette convertible 2012. |
55 | a photo of a chevrolet corvette zr1 2012. |
56 | a photo of a chevrolet corvette ron fellows edition z06 2007. |
57 | a photo of a chevrolet traverse suv 2012. |
58 | a photo of a chevrolet camaro convertible 2012. |
59 | a photo of a chevrolet hhr ss 2010. |
60 | a photo of a chevrolet impala sedan 2007. |
61 | a photo of a chevrolet tahoe hybrid suv 2012. |
62 | a photo of a chevrolet sonic sedan 2012. |
63 | a photo of a chevrolet express cargo van 2007. |
64 | a photo of a chevrolet avalanche crew cab 2012. |
65 | a photo of a chevrolet cobalt ss 2010. |
66 | a photo of a chevrolet malibu hybrid sedan 2010. |
67 | a photo of a chevrolet trailblazer ss 2009. |
68 | a photo of a chevrolet silverado 2500hd regular cab 2012. |
69 | a photo of a chevrolet silverado 1500 classic extended cab 2007. |
70 | a photo of a chevrolet express van 2007. |
71 | a photo of a chevrolet monte carlo coupe 2007. |
72 | a photo of a chevrolet malibu sedan 2007. |
73 | a photo of a chevrolet silverado 1500 extended cab 2012. |
74 | a photo of a chevrolet silverado 1500 regular cab 2012. |
75 | a photo of a chrysler aspen suv 2009. |
76 | a photo of a chrysler sebring convertible 2010. |
77 | a photo of a chrysler town and country minivan 2012. |
78 | a photo of a chrysler 300 srt-8 2010. |
79 | a photo of a chrysler crossfire convertible 2008. |
80 | a photo of a chrysler pt cruiser convertible 2008. |
81 | a photo of a daewoo nubira wagon 2002. |
82 | a photo of a dodge caliber wagon 2012. |
83 | a photo of a dodge caliber wagon 2007. |
84 | a photo of a dodge caravan minivan 1997. |
85 | a photo of a dodge ram pickup 3500 crew cab 2010. |
86 | a photo of a dodge ram pickup 3500 quad cab 2009. |
87 | a photo of a dodge sprinter cargo van 2009. |
88 | a photo of a dodge journey suv 2012. |
89 | a photo of a dodge dakota crew cab 2010. |
90 | a photo of a dodge dakota club cab 2007. |
91 | a photo of a dodge magnum wagon 2008. |
92 | a photo of a dodge challenger srt8 2011. |
93 | a photo of a dodge durango suv 2012. |
94 | a photo of a dodge durango suv 2007. |
95 | a photo of a dodge charger sedan 2012. |
96 | a photo of a dodge charger srt-8 2009. |
97 | a photo of a eagle talon hatchback 1998. |
98 | a photo of a fiat 500 abarth 2012. |
99 | a photo of a fiat 500 convertible 2012. |
End of preview. Expand in Data Studio
Classifying car images from 96 makes.
| Task category | ZeroShotClassification (image-to-text) |
| Domains | Scene |
| Reference | Collecting a Large-scale Dataset of Fine-grained Cars |
Source datasets:
How to evaluate on this task
You can evaluate an embedding model on this dataset using the following code:
import mteb
task = mteb.get_task("StanfordCarsZeroShot")
model = mteb.get_model(YOUR_MODEL)
mteb.evaluate(model, task)
To learn more about how to run models on mteb task check out the GitHub repository.
Citation
If you use this dataset, please cite the dataset as well as mteb, as this dataset likely includes additional processing as a part of the MMTEB Contribution.
@inproceedings{Krause2013CollectingAL,
author = {Jonathan Krause and Jia Deng and Michael Stark and Li Fei-Fei},
title = {Collecting a Large-scale Dataset of Fine-grained Cars},
url = {https://api.semanticscholar.org/CorpusID:16632981},
year = {2013},
}
@article{enevoldsen2025mmtebmassivemultilingualtext,
title={MMTEB: Massive Multilingual Text Embedding Benchmark},
author={Kenneth Enevoldsen and Isaac Chung and Imene Kerboua and Márton Kardos and Ashwin Mathur and David Stap and Jay Gala and Wissam Siblini and Dominik Krzemiński and Genta Indra Winata and Saba Sturua and Saiteja Utpala and Mathieu Ciancone and Marion Schaeffer and Gabriel Sequeira and Diganta Misra and Shreeya Dhakal and Jonathan Rystrøm and Roman Solomatin and Ömer Çağatan and Akash Kundu and Martin Bernstorff and Shitao Xiao and Akshita Sukhlecha and Bhavish Pahwa and Rafał Poświata and Kranthi Kiran GV and Shawon Ashraf and Daniel Auras and Björn Plüster and Jan Philipp Harries and Loïc Magne and Isabelle Mohr and Mariya Hendriksen and Dawei Zhu and Hippolyte Gisserot-Boukhlef and Tom Aarsen and Jan Kostkan and Konrad Wojtasik and Taemin Lee and Marek Šuppa and Crystina Zhang and Roberta Rocca and Mohammed Hamdy and Andrianos Michail and John Yang and Manuel Faysse and Aleksei Vatolin and Nandan Thakur and Manan Dey and Dipam Vasani and Pranjal Chitale and Simone Tedeschi and Nguyen Tai and Artem Snegirev and Michael Günther and Mengzhou Xia and Weijia Shi and Xing Han Lù and Jordan Clive and Gayatri Krishnakumar and Anna Maksimova and Silvan Wehrli and Maria Tikhonova and Henil Panchal and Aleksandr Abramov and Malte Ostendorff and Zheng Liu and Simon Clematide and Lester James Miranda and Alena Fenogenova and Guangyu Song and Ruqiya Bin Safi and Wen-Ding Li and Alessia Borghini and Federico Cassano and Hongjin Su and Jimmy Lin and Howard Yen and Lasse Hansen and Sara Hooker and Chenghao Xiao and Vaibhav Adlakha and Orion Weller and Siva Reddy and Niklas Muennighoff},
publisher = {arXiv},
journal={arXiv preprint arXiv:2502.13595},
year={2025},
url={https://arxiv.org/abs/2502.13595},
doi = {10.48550/arXiv.2502.13595},
}
@article{muennighoff2022mteb,
author = {Muennighoff, Niklas and Tazi, Nouamane and Magne, Loïc and Reimers, Nils},
title = {MTEB: Massive Text Embedding Benchmark},
publisher = {arXiv},
journal={arXiv preprint arXiv:2210.07316},
year = {2022}
url = {https://arxiv.org/abs/2210.07316},
doi = {10.48550/ARXIV.2210.07316},
}
Dataset Statistics
Dataset Statistics
The following code contains the descriptive statistics from the task. These can also be obtained using:
import mteb
task = mteb.get_task("StanfordCarsZeroShot")
desc_stats = task.metadata.descriptive_stats
{
"test": {
"num_samples": 8041,
"number_of_characters": null,
"text_statistics": null,
"image_statistics": {
"min_image_width": 78,
"average_image_width": 701.1770924014426,
"max_image_width": 7800,
"min_image_height": 41,
"average_image_height": 483.749658002736,
"max_image_height": 5400,
"unique_images": 8033
},
"label_statistics": {
"min_labels_per_text": 1,
"average_label_per_text": 1.0,
"max_labels_per_text": 1,
"unique_labels": 196,
"labels": {
"180": {
"count": 38
},
"102": {
"count": 39
},
"144": {
"count": 44
},
"186": {
"count": 43
},
"184": {
"count": 38
},
"77": {
"count": 37
},
"117": {
"count": 41
},
"164": {
"count": 44
},
"31": {
"count": 41
},
"59": {
"count": 36
},
"48": {
"count": 37
},
"107": {
"count": 44
},
"115": {
"count": 37
},
"134": {
"count": 42
},
"82": {
"count": 40
},
"50": {
"count": 43
},
"153": {
"count": 42
},
"32": {
"count": 42
},
"21": {
"count": 42
},
"150": {
"count": 43
},
"3": {
"count": 42
},
"80": {
"count": 45
},
"106": {
"count": 44
},
"190": {
"count": 46
},
"169": {
"count": 44
},
"194": {
"count": 43
},
"90": {
"count": 38
},
"4": {
"count": 40
},
"163": {
"count": 43
},
"147": {
"count": 45
},
"187": {
"count": 43
},
"43": {
"count": 44
},
"6": {
"count": 39
},
"30": {
"count": 44
},
"73": {
"count": 43
},
"29": {
"count": 41
},
"165": {
"count": 41
},
"179": {
"count": 42
},
"105": {
"count": 41
},
"2": {
"count": 43
},
"64": {
"count": 45
},
"34": {
"count": 41
},
"74": {
"count": 44
},
"84": {
"count": 43
},
"24": {
"count": 39
},
"167": {
"count": 42
},
"136": {
"count": 43
},
"133": {
"count": 33
},
"155": {
"count": 39
},
"119": {
"count": 42
},
"129": {
"count": 41
},
"127": {
"count": 39
},
"35": {
"count": 41
},
"170": {
"count": 46
},
"36": {
"count": 38
},
"63": {
"count": 29
},
"182": {
"count": 42
},
"42": {
"count": 46
},
"17": {
"count": 42
},
"75": {
"count": 43
},
"0": {
"count": 44
},
"62": {
"count": 44
},
"173": {
"count": 41
},
"16": {
"count": 40
},
"104": {
"count": 43
},
"49": {
"count": 42
},
"122": {
"count": 44
},
"81": {
"count": 45
},
"191": {
"count": 42
},
"92": {
"count": 39
},
"145": {
"count": 43
},
"95": {
"count": 41
},
"54": {
"count": 39
},
"114": {
"count": 45
},
"112": {
"count": 42
},
"151": {
"count": 35
},
"91": {
"count": 40
},
"188": {
"count": 40
},
"20": {
"count": 42
},
"33": {
"count": 44
},
"86": {
"count": 44
},
"128": {
"count": 38
},
"142": {
"count": 40
},
"19": {
"count": 46
},
"177": {
"count": 41
},
"11": {
"count": 36
},
"45": {
"count": 43
},
"60": {
"count": 43
},
"8": {
"count": 41
},
"56": {
"count": 37
},
"28": {
"count": 42
},
"120": {
"count": 44
},
"5": {
"count": 44
},
"85": {
"count": 42
},
"68": {
"count": 38
},
"22": {
"count": 39
},
"108": {
"count": 44
},
"89": {
"count": 41
},
"132": {
"count": 42
},
"125": {
"count": 42
},
"137": {
"count": 39
},
"158": {
"count": 36
},
"58": {
"count": 44
},
"123": {
"count": 39
},
"52": {
"count": 44
},
"27": {
"count": 41
},
"13": {
"count": 42
},
"70": {
"count": 35
},
"25": {
"count": 34
},
"185": {
"count": 38
},
"171": {
"count": 44
},
"9": {
"count": 33
},
"40": {
"count": 35
},
"178": {
"count": 45
},
"44": {
"count": 32
},
"97": {
"count": 46
},
"87": {
"count": 39
},
"159": {
"count": 44
},
"146": {
"count": 44
},
"51": {
"count": 41
},
"121": {
"count": 40
},
"1": {
"count": 32
},
"160": {
"count": 48
},
"78": {
"count": 48
},
"109": {
"count": 43
},
"103": {
"count": 42
},
"174": {
"count": 30
},
"181": {
"count": 46
},
"23": {
"count": 45
},
"111": {
"count": 45
},
"166": {
"count": 47
},
"172": {
"count": 43
},
"66": {
"count": 38
},
"192": {
"count": 41
},
"148": {
"count": 42
},
"72": {
"count": 44
},
"141": {
"count": 32
},
"71": {
"count": 45
},
"7": {
"count": 45
},
"152": {
"count": 44
},
"183": {
"count": 40
},
"98": {
"count": 27
},
"94": {
"count": 45
},
"126": {
"count": 41
},
"100": {
"count": 42
},
"131": {
"count": 43
},
"116": {
"count": 42
},
"39": {
"count": 39
},
"149": {
"count": 36
},
"101": {
"count": 39
},
"139": {
"count": 42
},
"69": {
"count": 42
},
"12": {
"count": 41
},
"14": {
"count": 43
},
"96": {
"count": 42
},
"41": {
"count": 34
},
"189": {
"count": 43
},
"10": {
"count": 38
},
"140": {
"count": 34
},
"26": {
"count": 35
},
"57": {
"count": 44
},
"88": {
"count": 44
},
"67": {
"count": 40
},
"93": {
"count": 43
},
"193": {
"count": 45
},
"161": {
"count": 45
},
"118": {
"count": 68
},
"110": {
"count": 42
},
"154": {
"count": 42
},
"138": {
"count": 42
},
"143": {
"count": 46
},
"61": {
"count": 37
},
"176": {
"count": 44
},
"113": {
"count": 45
},
"18": {
"count": 40
},
"53": {
"count": 40
},
"47": {
"count": 42
},
"157": {
"count": 29
},
"168": {
"count": 38
},
"124": {
"count": 43
},
"79": {
"count": 43
},
"130": {
"count": 42
},
"46": {
"count": 35
},
"55": {
"count": 46
},
"195": {
"count": 40
},
"38": {
"count": 36
},
"37": {
"count": 40
},
"99": {
"count": 33
},
"83": {
"count": 42
},
"162": {
"count": 36
},
"135": {
"count": 24
},
"175": {
"count": 38
},
"156": {
"count": 36
},
"15": {
"count": 43
},
"65": {
"count": 41
},
"76": {
"count": 40
}
}
},
"candidates_labels_text_statistics": {
"total_text_length": 8002,
"min_text_length": 29,
"average_text_length": 40.826530612244895,
"max_text_length": 68,
"unique_texts": 196
}
}
}
This dataset card was automatically generated using MTEB
- Downloads last month
- 72