BAAI

/

bge-m3

like 815

Thanks for your interest in our work! The current MTEB focuses on single language retrieval (English/Chinese), while the goal of bge-m3 is versatility(Multi-Functionality, Multi-Linguality, and Multi-Granularity),
so we didn't report the performance of bge-m3 on it.
We are working on testing the performance of bge-m3 on more datasets, and will release these results as soon as possible.

Ran around 50% of MTEB, here are some scores for orientation.

tags:
- mteb
model-index:
- name: bge-m3
results:
- task:
type: Classification
dataset:
type: mteb/amazon_counterfactual
name: MTEB AmazonCounterfactualClassification (en)
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- type: accuracy
value: 75.61194029850746
- type: ap
value: 39.4441857734361
- type: f1
value: 69.97034098944273
- task:
type: Classification
dataset:
type: mteb/amazon_polarity
name: MTEB AmazonPolarityClassification
config: default
split: test
revision: e2d317d38cd51312af73b3d32a06d1a08b442046
metrics:
- type: accuracy
value: 91.005775
- type: ap
value: 87.2965129654928
- type: f1
value: 90.99368816155456
- task:
type: Classification
dataset:
type: mteb/amazon_reviews_multi
name: MTEB AmazonReviewsClassification (en)
config: en
split: test
revision: 1399c76144fd37290681b995c656ef9b2e06e26d
metrics:
- type: accuracy
value: 46.976
- type: f1
value: 44.92055541556327
- task:
type: Retrieval
dataset:
type: arguana
name: MTEB ArguAna
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 28.733999999999998
- type: map_at_10
value: 45.134
- type: map_at_100
value: 46.04
- type: map_at_1000
value: 46.047
- type: map_at_3
value: 40.149
- type: map_at_5
value: 43.24
- type: mrr_at_1
value: 29.587000000000003
- type: mrr_at_10
value: 45.428000000000004
- type: mrr_at_100
value: 46.336
- type: mrr_at_1000
value: 46.343
- type: mrr_at_3
value: 40.505
- type: mrr_at_5
value: 43.531
- type: ndcg_at_1
value: 28.733999999999998
- type: ndcg_at_10
value: 53.994
- type: ndcg_at_100
value: 57.742000000000004
- type: ndcg_at_1000
value: 57.908
- type: ndcg_at_3
value: 43.901
- type: ndcg_at_5
value: 49.437999999999995
- type: precision_at_1
value: 28.733999999999998
- type: precision_at_10
value: 8.215
- type: precision_at_100
value: 0.9820000000000001
- type: precision_at_1000
value: 0.1
- type: precision_at_3
value: 18.255
- type: precision_at_5
value: 13.627
- type: recall_at_1
value: 28.733999999999998
- type: recall_at_10
value: 82.148
- type: recall_at_100
value: 98.222
- type: recall_at_1000
value: 99.502
- type: recall_at_3
value: 54.765
- type: recall_at_5
value: 68.137
- task:
type: Clustering
dataset:
type: mteb/arxiv-clustering-p2p
name: MTEB ArxivClusteringP2P
config: default
split: test
revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d
metrics:
- type: v_measure
value: 39.418453651259746
- task:
type: Clustering
dataset:
type: mteb/arxiv-clustering-s2s
name: MTEB ArxivClusteringS2S
config: default
split: test
revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53
metrics:
- type: v_measure
value: 27.769780397780057
- task:
type: Reranking
dataset:
type: mteb/askubuntudupquestions-reranking
name: MTEB AskUbuntuDupQuestions
config: default
split: test
revision: 2000358ca161889fa9c082cb41daa8dcfb161a54
metrics:
- type: map
value: 61.47747150445644
- type: mrr
value: 74.28241656773513
- task:
type: BitextMining
dataset:
type: mteb/bucc-bitext-mining
name: MTEB BUCC (de-en)
config: de-en
split: test
revision: d51519689f32196a32af33b075a01d0e7c51e252
metrics:
- type: accuracy
value: 99.53027139874739
- type: f1
value: 99.45720250521921
- type: precision
value: 99.42066805845512
- type: recall
value: 99.53027139874739
- task:
type: BitextMining
dataset:
type: mteb/bucc-bitext-mining
name: MTEB BUCC (fr-en)
config: fr-en
split: test
revision: d51519689f32196a32af33b075a01d0e7c51e252
metrics:
- type: accuracy
value: 98.72331058771738
- type: f1
value: 98.55088414410449
- type: precision
value: 98.46650524616628
- type: recall
value: 98.72331058771738
- task:
type: BitextMining
dataset:
type: mteb/bucc-bitext-mining
name: MTEB BUCC (ru-en)
config: ru-en
split: test
revision: d51519689f32196a32af33b075a01d0e7c51e252
metrics:
- type: accuracy
value: 97.90093522687911
- type: f1
value: 97.64923219027827
- type: precision
value: 97.52338067197783
- type: recall
value: 97.90093522687911
- task:
type: BitextMining
dataset:
type: mteb/bucc-bitext-mining
name: MTEB BUCC (zh-en)
config: zh-en
split: test
revision: d51519689f32196a32af33b075a01d0e7c51e252
metrics:
- type: accuracy
value: 99.0521327014218
- type: f1
value: 98.99947340705634
- type: precision
value: 98.97314375987362
- type: recall
value: 99.0521327014218
- task:
type: Classification
dataset:
type: mteb/banking77
name: MTEB Banking77Classification
config: default
split: test
revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
metrics:
- type: accuracy
value: 81.93181818181819
- type: f1
value: 81.0852312152688
- task:
type: Clustering
dataset:
type: mteb/biorxiv-clustering-p2p
name: MTEB BiorxivClusteringP2P
config: default
split: test
revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40
metrics:
- type: v_measure
value: 32.856148672160444
- task:
type: Clustering
dataset:
type: mteb/biorxiv-clustering-s2s
name: MTEB BiorxivClusteringS2S
config: default
split: test
revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908
metrics:
- type: v_measure
value: 25.729734227390967
- task:
type: BitextMining
dataset:
type: strombergnlp/bornholmsk_parallel
name: MTEB BornholmBitextMining
config: default
split: test
revision: 3bc5cfb4ec514264fe2db5615fac9016f7251552
metrics:
- type: accuracy
value: 52.6
- type: f1
value: 44.05714285714286
- type: precision
value: 40.93380952380952
- type: recall
value: 52.6
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackAndroidRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 31.1
- type: map_at_10
value: 42.028
- type: map_at_100
value: 43.57
- type: map_at_1000
value: 43.683
- type: map_at_3
value: 38.479
- type: map_at_5
value: 40.502
- type: mrr_at_1
value: 38.054
- type: mrr_at_10
value: 48.073
- type: mrr_at_100
value: 48.824
- type: mrr_at_1000
value: 48.856
- type: mrr_at_3
value: 45.231
- type: mrr_at_5
value: 46.941
- type: ndcg_at_1
value: 37.911
- type: ndcg_at_10
value: 48.406
- type: ndcg_at_100
value: 54.057
- type: ndcg_at_1000
value: 55.81700000000001
- type: ndcg_at_3
value: 42.988
- type: ndcg_at_5
value: 45.686
- type: precision_at_1
value: 37.911
- type: precision_at_10
value: 9.285
- type: precision_at_100
value: 1.498
- type: precision_at_1000
value: 0.196
- type: precision_at_3
value: 20.41
- type: precision_at_5
value: 14.907
- type: recall_at_1
value: 31.1
- type: recall_at_10
value: 60.50299999999999
- type: recall_at_100
value: 84.313
- type: recall_at_1000
value: 95.44500000000001
- type: recall_at_3
value: 45.550000000000004
- type: recall_at_5
value: 52.735
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackEnglishRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 28.720000000000002
- type: map_at_10
value: 38.87
- type: map_at_100
value: 40.141
- type: map_at_1000
value: 40.27
- type: map_at_3
value: 36.064
- type: map_at_5
value: 37.659
- type: mrr_at_1
value: 35.86
- type: mrr_at_10
value: 44.458
- type: mrr_at_100
value: 45.221000000000004
- type: mrr_at_1000
value: 45.266
- type: mrr_at_3
value: 42.335
- type: mrr_at_5
value: 43.59
- type: ndcg_at_1
value: 35.86
- type: ndcg_at_10
value: 44.397
- type: ndcg_at_100
value: 48.924
- type: ndcg_at_1000
value: 50.949
- type: ndcg_at_3
value: 40.382
- type: ndcg_at_5
value: 42.308
- type: precision_at_1
value: 35.86
- type: precision_at_10
value: 8.369
- type: precision_at_100
value: 1.369
- type: precision_at_1000
value: 0.181
- type: precision_at_3
value: 19.448
- type: precision_at_5
value: 13.783000000000001
- type: recall_at_1
value: 28.720000000000002
- type: recall_at_10
value: 54.057
- type: recall_at_100
value: 73.015
- type: recall_at_1000
value: 85.90100000000001
- type: recall_at_3
value: 42.561
- type: recall_at_5
value: 47.827999999999996
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackGamingRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 37.798
- type: map_at_10
value: 49.559999999999995
- type: map_at_100
value: 50.586
- type: map_at_1000
value: 50.64300000000001
- type: map_at_3
value: 46.686
- type: map_at_5
value: 48.292
- type: mrr_at_1
value: 42.884
- type: mrr_at_10
value: 52.864999999999995
- type: mrr_at_100
value: 53.549
- type: mrr_at_1000
value: 53.583000000000006
- type: mrr_at_3
value: 50.658
- type: mrr_at_5
value: 51.834
- type: ndcg_at_1
value: 42.884
- type: ndcg_at_10
value: 55.147
- type: ndcg_at_100
value: 59.205
- type: ndcg_at_1000
value: 60.431999999999995
- type: ndcg_at_3
value: 50.268
- type: ndcg_at_5
value: 52.595000000000006
- type: precision_at_1
value: 42.884
- type: precision_at_10
value: 8.802999999999999
- type: precision_at_100
value: 1.1769999999999998
- type: precision_at_1000
value: 0.133
- type: precision_at_3
value: 22.299
- type: precision_at_5
value: 15.160000000000002
- type: recall_at_1
value: 37.798
- type: recall_at_10
value: 68.28
- type: recall_at_100
value: 85.697
- type: recall_at_1000
value: 94.447
- type: recall_at_3
value: 55.184999999999995
- type: recall_at_5
value: 60.940000000000005
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackGisRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 24.935
- type: map_at_10
value: 32.757999999999996
- type: map_at_100
value: 33.633
- type: map_at_1000
value: 33.716
- type: map_at_3
value: 30.168
- type: map_at_5
value: 31.513999999999996
- type: mrr_at_1
value: 27.119
- type: mrr_at_10
value: 34.885
- type: mrr_at_100
value: 35.674
- type: mrr_at_1000
value: 35.739
- type: mrr_at_3
value: 32.505
- type: mrr_at_5
value: 33.714
- type: ndcg_at_1
value: 27.006000000000004
- type: ndcg_at_10
value: 37.56
- type: ndcg_at_100
value: 42.025
- type: ndcg_at_1000
value: 44.297
- type: ndcg_at_3
value: 32.408
- type: ndcg_at_5
value: 34.64
- type: precision_at_1
value: 27.006000000000004
- type: precision_at_10
value: 5.774
- type: precision_at_100
value: 0.844
- type: precision_at_1000
value: 0.108
- type: precision_at_3
value: 13.635
- type: precision_at_5
value: 9.424000000000001
- type: recall_at_1
value: 24.935
- type: recall_at_10
value: 50.608
- type: recall_at_100
value: 71.344
- type: recall_at_1000
value: 88.67699999999999
- type: recall_at_3
value: 36.368
- type: recall_at_5
value: 41.81
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackMathematicaRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 15.723
- type: map_at_10
value: 22.159000000000002
- type: map_at_100
value: 23.399
- type: map_at_1000
value: 23.513
- type: map_at_3
value: 19.792
- type: map_at_5
value: 21.102999999999998
- type: mrr_at_1
value: 19.527
- type: mrr_at_10
value: 26.346999999999998
- type: mrr_at_100
value: 27.48
- type: mrr_at_1000
value: 27.544999999999998
- type: mrr_at_3
value: 23.983999999999998
- type: mrr_at_5
value: 25.178
- type: ndcg_at_1
value: 19.403000000000002
- type: ndcg_at_10
value: 26.789
- type: ndcg_at_100
value: 32.791
- type: ndcg_at_1000
value: 35.58
- type: ndcg_at_3
value: 22.235
- type: ndcg_at_5
value: 24.27
- type: precision_at_1
value: 19.403000000000002
- type: precision_at_10
value: 4.913
- type: precision_at_100
value: 0.903
- type: precision_at_1000
value: 0.128
- type: precision_at_3
value: 10.488999999999999
- type: precision_at_5
value: 7.710999999999999
- type: recall_at_1
value: 15.723
- type: recall_at_10
value: 37.143
- type: recall_at_100
value: 63.402
- type: recall_at_1000
value: 83.291
- type: recall_at_3
value: 24.446
- type: recall_at_5
value: 29.544999999999998
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackPhysicsRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 26.606
- type: map_at_10
value: 36.495
- type: map_at_100
value: 37.759
- type: map_at_1000
value: 37.877
- type: map_at_3
value: 33.542
- type: map_at_5
value: 35.256
- type: mrr_at_1
value: 32.435
- type: mrr_at_10
value: 42.126999999999995
- type: mrr_at_100
value: 42.96
- type: mrr_at_1000
value: 43.006
- type: mrr_at_3
value: 39.75
- type: mrr_at_5
value: 41.169
- type: ndcg_at_1
value: 32.531
- type: ndcg_at_10
value: 42.382
- type: ndcg_at_100
value: 47.721000000000004
- type: ndcg_at_1000
value: 49.888
- type: ndcg_at_3
value: 37.687
- type: ndcg_at_5
value: 39.987
- type: precision_at_1
value: 32.531
- type: precision_at_10
value: 7.7
- type: precision_at_100
value: 1.209
- type: precision_at_1000
value: 0.159
- type: precision_at_3
value: 18.062
- type: precision_at_5
value: 12.723999999999998
- type: recall_at_1
value: 26.606
- type: recall_at_10
value: 54.277
- type: recall_at_100
value: 76.91
- type: recall_at_1000
value: 90.979
- type: recall_at_3
value: 40.692
- type: recall_at_5
value: 46.949999999999996
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackProgrammersRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 25.306
- type: map_at_10
value: 34.676
- type: map_at_100
value: 35.939
- type: map_at_1000
value: 36.044
- type: map_at_3
value: 31.418000000000003
- type: map_at_5
value: 33.323
- type: mrr_at_1
value: 31.507
- type: mrr_at_10
value: 39.893
- type: mrr_at_100
value: 40.808
- type: mrr_at_1000
value: 40.859
- type: mrr_at_3
value: 37.080999999999996
- type: mrr_at_5
value: 38.708
- type: ndcg_at_1
value: 31.392999999999997
- type: ndcg_at_10
value: 40.357
- type: ndcg_at_100
value: 45.878
- type: ndcg_at_1000
value: 48.101
- type: ndcg_at_3
value: 34.914
- type: ndcg_at_5
value: 37.476
- type: precision_at_1
value: 31.392999999999997
- type: precision_at_10
value: 7.352
- type: precision_at_100
value: 1.1769999999999998
- type: precision_at_1000
value: 0.153
- type: precision_at_3
value: 16.476
- type: precision_at_5
value: 11.918
- type: recall_at_1
value: 25.306
- type: recall_at_10
value: 52.528
- type: recall_at_100
value: 76.069
- type: recall_at_1000
value: 91.106
- type: recall_at_3
value: 37.275000000000006
- type: recall_at_5
value: 44.080999999999996
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackStatsRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 23.159
- type: map_at_10
value: 29.647000000000002
- type: map_at_100
value: 30.441000000000003
- type: map_at_1000
value: 30.531999999999996
- type: map_at_3
value: 27.076
- type: map_at_5
value: 28.792
- type: mrr_at_1
value: 25.613000000000003
- type: mrr_at_10
value: 32.123000000000005
- type: mrr_at_100
value: 32.814
- type: mrr_at_1000
value: 32.883
- type: mrr_at_3
value: 29.755
- type: mrr_at_5
value: 31.296000000000003
- type: ndcg_at_1
value: 25.613000000000003
- type: ndcg_at_10
value: 33.707
- type: ndcg_at_100
value: 37.62
- type: ndcg_at_1000
value: 40.042
- type: ndcg_at_3
value: 29.086000000000002
- type: ndcg_at_5
value: 31.823
- type: precision_at_1
value: 25.613000000000003
- type: precision_at_10
value: 5.2909999999999995
- type: precision_at_100
value: 0.7849999999999999
- type: precision_at_1000
value: 0.107
- type: precision_at_3
value: 12.27
- type: precision_at_5
value: 9.11
- type: recall_at_1
value: 23.159
- type: recall_at_10
value: 43.795
- type: recall_at_100
value: 61.436
- type: recall_at_1000
value: 79.511
- type: recall_at_3
value: 31.394
- type: recall_at_5
value: 38.002
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackTexRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 17.523
- type: map_at_10
value: 24.34
- type: map_at_100
value: 25.318
- type: map_at_1000
value: 25.432
- type: map_at_3
value: 22.31
- type: map_at_5
value: 23.406
- type: mrr_at_1
value: 21.163
- type: mrr_at_10
value: 27.907
- type: mrr_at_100
value: 28.816000000000003
- type: mrr_at_1000
value: 28.89
- type: mrr_at_3
value: 26.015
- type: mrr_at_5
value: 27.046
- type: ndcg_at_1
value: 21.129
- type: ndcg_at_10
value: 28.542
- type: ndcg_at_100
value: 33.465
- type: ndcg_at_1000
value: 36.339
- type: ndcg_at_3
value: 24.905
- type: ndcg_at_5
value: 26.489
- type: precision_at_1
value: 21.129
- type: precision_at_10
value: 5.045
- type: precision_at_100
value: 0.8699999999999999
- type: precision_at_1000
value: 0.129
- type: precision_at_3
value: 11.643
- type: precision_at_5
value: 8.245
- type: recall_at_1
value: 17.523
- type: recall_at_10
value: 37.682
- type: recall_at_100
value: 60.111000000000004
- type: recall_at_1000
value: 80.766
- type: recall_at_3
value: 27.412
- type: recall_at_5
value: 31.518
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackUnixRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 26.284000000000002
- type: map_at_10
value: 34.435
- type: map_at_100
value: 35.545
- type: map_at_1000
value: 35.644
- type: map_at_3
value: 32.036
- type: map_at_5
value: 33.335
- type: mrr_at_1
value: 30.784
- type: mrr_at_10
value: 38.261
- type: mrr_at_100
value: 39.153
- type: mrr_at_1000
value: 39.208999999999996
- type: mrr_at_3
value: 36.022999999999996
- type: mrr_at_5
value: 37.282
- type: ndcg_at_1
value: 30.784
- type: ndcg_at_10
value: 39.091
- type: ndcg_at_100
value: 44.391999999999996
- type: ndcg_at_1000
value: 46.681
- type: ndcg_at_3
value: 34.778
- type: ndcg_at_5
value: 36.664
- type: precision_at_1
value: 30.784
- type: precision_at_10
value: 6.446000000000001
- type: precision_at_100
value: 1.014
- type: precision_at_1000
value: 0.132
- type: precision_at_3
value: 15.672
- type: precision_at_5
value: 10.84
- type: recall_at_1
value: 26.284000000000002
- type: recall_at_10
value: 49.467
- type: recall_at_100
value: 73.039
- type: recall_at_1000
value: 89.037
- type: recall_at_3
value: 37.519999999999996
- type: recall_at_5
value: 42.435
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackWebmastersRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 25.263999999999996
- type: map_at_10
value: 34.665
- type: map_at_100
value: 36.137
- type: map_at_1000
value: 36.354
- type: map_at_3
value: 32.137
- type: map_at_5
value: 33.371
- type: mrr_at_1
value: 31.225
- type: mrr_at_10
value: 39.948
- type: mrr_at_100
value: 40.794999999999995
- type: mrr_at_1000
value: 40.858
- type: mrr_at_3
value: 37.879000000000005
- type: mrr_at_5
value: 39.035
- type: ndcg_at_1
value: 30.830000000000002
- type: ndcg_at_10
value: 40.516000000000005
- type: ndcg_at_100
value: 45.7
- type: ndcg_at_1000
value: 48.446
- type: ndcg_at_3
value: 36.603
- type: ndcg_at_5
value: 38.098
- type: precision_at_1
value: 30.830000000000002
- type: precision_at_10
value: 7.767
- type: precision_at_100
value: 1.48
- type: precision_at_1000
value: 0.23800000000000002
- type: precision_at_3
value: 17.391000000000002
- type: precision_at_5
value: 12.292
- type: recall_at_1
value: 25.263999999999996
- type: recall_at_10
value: 51.034
- type: recall_at_100
value: 74.493
- type: recall_at_1000
value: 92.15100000000001
- type: recall_at_3
value: 39.519999999999996
- type: recall_at_5
value: 43.74
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackWordpressRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 20.4
- type: map_at_10
value: 27.72
- type: map_at_100
value: 28.57
- type: map_at_1000
value: 28.682999999999996
- type: map_at_3
value: 25.365
- type: map_at_5
value: 26.773000000000003
- type: mrr_at_1
value: 22.551
- type: mrr_at_10
value: 29.829
- type: mrr_at_100
value: 30.604
- type: mrr_at_1000
value: 30.691000000000003
- type: mrr_at_3
value: 27.695999999999998
- type: mrr_at_5
value: 28.962
- type: ndcg_at_1
value: 22.551
- type: ndcg_at_10
value: 32.096999999999994
- type: ndcg_at_100
value: 36.518
- type: ndcg_at_1000
value: 39.487
- type: ndcg_at_3
value: 27.655
- type: ndcg_at_5
value: 29.975
- type: precision_at_1
value: 22.551
- type: precision_at_10
value: 5.083
- type: precision_at_100
value: 0.787
- type: precision_at_1000
value: 0.117
- type: precision_at_3
value: 11.83
- type: precision_at_5
value: 8.577
- type: recall_at_1
value: 20.4
- type: recall_at_10
value: 43.444
- type: recall_at_100
value: 64.19500000000001
- type: recall_at_1000
value: 86.586
- type: recall_at_3
value: 31.75
- type: recall_at_5
value: 37.214999999999996
- task:
type: Retrieval
dataset:
type: dbpedia-entity
name: MTEB DBPedia
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 8.709
- type: map_at_10
value: 18.137
- type: map_at_100
value: 25.337
- type: map_at_1000
value: 26.929
- type: map_at_3
value: 13.164000000000001
- type: map_at_5
value: 15.332
- type: mrr_at_1
value: 63.5
- type: mrr_at_10
value: 72.938
- type: mrr_at_100
value: 73.248
- type: mrr_at_1000
value: 73.25500000000001
- type: mrr_at_3
value: 70.917
- type: mrr_at_5
value: 72.317
- type: ndcg_at_1
value: 52.75
- type: ndcg_at_10
value: 39.837
- type: ndcg_at_100
value: 44.204
- type: ndcg_at_1000
value: 51.514
- type: ndcg_at_3
value: 44.178
- type: ndcg_at_5
value: 41.703
- type: precision_at_1
value: 63.5
- type: precision_at_10
value: 31.5
- type: precision_at_100
value: 10.152999999999999
- type: precision_at_1000
value: 1.9929999999999999
- type: precision_at_3
value: 46.916999999999994
- type: precision_at_5
value: 39.85
- type: recall_at_1
value: 8.709
- type: recall_at_10
value: 23.595
- type: recall_at_100
value: 50.244
- type: recall_at_1000
value: 73.193
- type: recall_at_3
value: 14.504
- type: recall_at_5
value: 18.137
- task:
type: Classification
dataset:
type: mteb/emotion
name: MTEB EmotionClassification
config: default
split: test
revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37
metrics:
- type: accuracy
value: 50.154999999999994
- type: f1
value: 43.57245108706806
- task:
type: Retrieval
dataset:
type: fiqa
name: MTEB FiQA2018
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 20.968999999999998
- type: map_at_10
value: 33.664
- type: map_at_100
value: 35.471000000000004
- type: map_at_1000
value: 35.65
- type: map_at_3
value: 29.997
- type: map_at_5
value: 31.921
- type: mrr_at_1
value: 42.13
- type: mrr_at_10
value: 50.475
- type: mrr_at_100
value: 51.23799999999999
- type: mrr_at_1000
value: 51.281
- type: mrr_at_3
value: 48.687999999999995
- type: mrr_at_5
value: 49.591
- type: ndcg_at_1
value: 42.284
- type: ndcg_at_10
value: 41.365
- type: ndcg_at_100
value: 47.863
- type: ndcg_at_1000
value: 50.971999999999994
- type: ndcg_at_3
value: 38.84
- type: ndcg_at_5
value: 39.184000000000005
- type: precision_at_1
value: 42.284
- type: precision_at_10
value: 11.265
- type: precision_at_100
value: 1.7850000000000001
- type: precision_at_1000
value: 0.23500000000000001
- type: precision_at_3
value: 25.977
- type: precision_at_5
value: 18.426000000000002
- type: recall_at_1
value: 20.968999999999998
- type: recall_at_10
value: 47.239
- type: recall_at_100
value: 71.334
- type: recall_at_1000
value: 89.916
- type: recall_at_3
value: 35.38
- type: recall_at_5
value: 39.965
- task:
type: Classification
dataset:
type: mteb/imdb
name: MTEB ImdbClassification
config: default
split: test
revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7
metrics:
- type: accuracy
value: 87.80999999999999
- type: ap
value: 82.64452124367875
- type: f1
value: 87.79250775511873
- task:
type: Classification
dataset:
type: mteb/mtop_domain
name: MTEB MTOPDomainClassification (en)
config: en
split: test
revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
metrics:
- type: accuracy
value: 93.3766529867761
- type: f1
value: 93.07516766462871
- task:
type: Classification
dataset:
type: mteb/mtop_intent
name: MTEB MTOPIntentClassification (en)
config: en
split: test
revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
metrics:
- type: accuracy
value: 66.59598723210213
- type: f1
value: 47.047272522163624
- task:
type: Classification
dataset:
type: mteb/amazon_massive_intent
name: MTEB MassiveIntentClassification (en)
config: en
split: test
revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
metrics:
- type: accuracy
value: 71.06926698049764
- type: f1
value: 69.13014185231128
- task:
type: Classification
dataset:
type: mteb/amazon_massive_scenario
name: MTEB MassiveScenarioClassification (en)
config: en
split: test
revision: 7d571f92784cd94a019292a1f45445077d0ef634
metrics:
- type: accuracy
value: 76.66442501681236
- type: f1
value: 75.63836848136394
- task:
type: Clustering
dataset:
type: mteb/medrxiv-clustering-p2p
name: MTEB MedrxivClusteringP2P
config: default
split: test
revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73
metrics:
- type: v_measure
value: 30.23749152164288
- task:
type: Clustering
dataset:
type: mteb/medrxiv-clustering-s2s
name: MTEB MedrxivClusteringS2S
config: default
split: test
revision: 35191c8c0dca72d8ff3efcd72aa802307d469663
metrics:
- type: v_measure
value: 26.083238688599486
- task:
type: Retrieval
dataset:
type: nfcorpus
name: MTEB NFCorpus
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 5.272
- type: map_at_10
value: 11.716999999999999
- type: map_at_100
value: 14.402999999999999
- type: map_at_1000
value: 15.664
- type: map_at_3
value: 8.956
- type: map_at_5
value: 10.249
- type: mrr_at_1
value: 43.963
- type: mrr_at_10
value: 52.553000000000004
- type: mrr_at_100
value: 53.176
- type: mrr_at_1000
value: 53.222
- type: mrr_at_3
value: 51.135
- type: mrr_at_5
value: 51.94
- type: ndcg_at_1
value: 41.486000000000004
- type: ndcg_at_10
value: 31.342
- type: ndcg_at_100
value: 28.471999999999998
- type: ndcg_at_1000
value: 37.102000000000004
- type: ndcg_at_3
value: 37.736
- type: ndcg_at_5
value: 35.105
- type: precision_at_1
value: 43.344
- type: precision_at_10
value: 22.786
- type: precision_at_100
value: 7.093000000000001
- type: precision_at_1000
value: 1.9539999999999997
- type: precision_at_3
value: 35.707
- type: precision_at_5
value: 30.217
- type: recall_at_1
value: 5.272
- type: recall_at_10
value: 14.671999999999999
- type: recall_at_100
value: 28.308
- type: recall_at_1000
value: 59.11300000000001
- type: recall_at_3
value: 10.07
- type: recall_at_5
value: 11.99
- task:
type: Clustering
dataset:
type: mteb/reddit-clustering
name: MTEB RedditClustering
config: default
split: test
revision: 24640382cdbf8abc73003fb0fa6d111a705499eb
metrics:
- type: v_measure
value: 45.473096154259935
- task:
type: Clustering
dataset:
type: mteb/reddit-clustering-p2p
name: MTEB RedditClusteringP2P
config: default
split: test
revision: 282350215ef01743dc01b456c7f5241fa8937f16
metrics:
- type: v_measure
value: 57.54220073389692
- task:
type: Reranking
dataset:
type: mteb/scidocs-reranking
name: MTEB SciDocsRR
config: default
split: test
revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab
metrics:
- type: map
value: 77.38133262065693
- type: mrr
value: 93.20244479558205
- task:
type: PairClassification
dataset:
type: mteb/sprintduplicatequestions-pairclassification
name: MTEB SprintDuplicateQuestions
config: default
split: test
revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
metrics:
- type: cos_sim_accuracy
value: 99.87722772277228
- type: cos_sim_ap
value: 97.32321748519148
- type: cos_sim_f1
value: 93.74369323915236
- type: cos_sim_precision
value: 94.60285132382892
- type: cos_sim_recall
value: 92.9
- type: dot_accuracy
value: 99.87722772277228
- type: dot_ap
value: 97.31598992113545
- type: dot_f1
value: 93.74369323915236
- type: dot_precision
value: 94.60285132382892
- type: dot_recall
value: 92.9
- type: euclidean_accuracy
value: 99.87722772277228
- type: euclidean_ap
value: 97.32307655128027
- type: euclidean_f1
value: 93.74369323915236
- type: euclidean_precision
value: 94.60285132382892
- type: euclidean_recall
value: 92.9
- type: manhattan_accuracy
value: 99.87524752475248
- type: manhattan_ap
value: 97.2914492164419
- type: manhattan_f1
value: 93.58074222668003
- type: manhattan_precision
value: 93.86317907444668
- type: manhattan_recall
value: 93.30000000000001
- type: max_accuracy
value: 99.87722772277228
- type: max_ap
value: 97.32321748519148
- type: max_f1
value: 93.74369323915236
- task:
type: Clustering
dataset:
type: mteb/stackexchange-clustering
name: MTEB StackExchangeClustering
config: default
split: test
revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259
metrics:
- type: v_measure
value: 55.615368500858054
- task:
type: Clustering
dataset:
type: mteb/stackexchange-clustering-p2p
name: MTEB StackExchangeClusteringP2P
config: default
split: test
revision: 815ca46b2622cec33ccafc3735d572c266efdb44
metrics:
- type: v_measure
value: 31.827358732957318
- task:
type: Reranking
dataset:
type: mteb/stackoverflowdupquestions-reranking
name: MTEB StackOverflowDupQuestions
config: default
split: test
revision: e185fbe320c72810689fc5848eb6114e1ef5ec69
metrics:
- type: map
value: 51.58195276065337
- type: mrr
value: 52.538900029708856
- task:
type: Classification
dataset:
type: mteb/toxic_conversations_50k
name: MTEB ToxicConversationsClassification
config: default
split: test
revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c
metrics:
- type: accuracy
value: 72.5814
- type: ap
value: 15.705026098892539
- type: f1
value: 56.287318053086665
- task:
type: Classification
dataset:
type: mteb/tweet_sentiment_extraction
name: MTEB TweetSentimentExtractionClassification
config: default
split: test
revision: d604517c81ca91fe16a244d1248fc021f9ecee7a
metrics:
- type: accuracy
value: 63.7040181097906
- type: f1
value: 63.536790312918676
- task:
type: Clustering
dataset:
type: mteb/twentynewsgroups-clustering
name: MTEB TwentyNewsgroupsClustering
config: default
split: test
revision: 6125ec4e24fa026cec8a478383ee943acfbd5449
metrics:
- type: v_measure
value: 37.0755355816832
- task:
type: PairClassification
dataset:
type: mteb/twittersemeval2015-pairclassification
name: MTEB TwitterSemEval2015
config: default
split: test
revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
metrics:
- type: cos_sim_accuracy
value: 85.09268641592656
- type: cos_sim_ap
value: 70.26996907384613
- type: cos_sim_f1
value: 65.01384676249505
- type: cos_sim_precision
value: 64.98813604007381
- type: cos_sim_recall
value: 65.03957783641161
- type: dot_accuracy
value: 85.09268641592656
- type: dot_ap
value: 70.21842436196593
- type: dot_f1
value: 64.98002663115845
- type: dot_precision
value: 65.59139784946237
- type: dot_recall
value: 64.37994722955145
- type: euclidean_accuracy
value: 85.09268641592656
- type: euclidean_ap
value: 70.2701643775076
- type: euclidean_f1
value: 65.02177068214804
- type: euclidean_precision
value: 65.03035101609923
- type: euclidean_recall
value: 65.0131926121372
- type: manhattan_accuracy
value: 85.027120462538
- type: manhattan_ap
value: 70.04080344828975
- type: manhattan_f1
value: 64.59036898061288
- type: manhattan_precision
value: 61.40309155766944
- type: manhattan_recall
value: 68.12664907651715
- type: max_accuracy
value: 85.09268641592656
- type: max_ap
value: 70.2701643775076
- type: max_f1
value: 65.02177068214804
- task:
type: PairClassification
dataset:
type: mteb/twitterurlcorpus-pairclassification
name: MTEB TwitterURLCorpus
config: default
split: test
revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
metrics:
- type: cos_sim_accuracy
value: 88.99949547871309
- type: cos_sim_ap
value: 85.82630890007624
- type: cos_sim_f1
value: 78.36594626168225
- type: cos_sim_precision
value: 74.51402388225493
- type: cos_sim_recall
value: 82.63781952571605
- type: dot_accuracy
value: 88.99561454573679
- type: dot_ap
value: 85.80051087301636
- type: dot_f1
value: 78.36163544931128
- type: dot_precision
value: 74.38771274387713
- type: dot_recall
value: 82.78410840776101
- type: euclidean_accuracy
value: 89.00531687817751
- type: euclidean_ap
value: 85.82673451926944
- type: euclidean_f1
value: 78.36308545978899
- type: euclidean_precision
value: 74.50885109337035
- type: euclidean_recall
value: 82.63781952571605
- type: manhattan_accuracy
value: 88.9781503473435
- type: manhattan_ap
value: 85.7726134330323
- type: manhattan_f1
value: 78.3766878840281
- type: manhattan_precision
value: 75.05461207807765
- type: manhattan_recall
value: 82.00646750846936
- type: max_accuracy
value: 89.00531687817751
- type: max_ap
value: 85.82673451926944
- type: max_f1
value: 78.3766878840281

Thanks @michaelfeil for running the benchmarks. Looks promising! I think the main question that many ask themselves with FlagEmbeddings is whether to use https://huggingface.co/BAAI/bge-base-en-v1.5 or the newer https://huggingface.co/BAAI/bge-m3 as it's really versatile, multilingual and has a larger context. @Shitao do you maybe have general recommendations here what model to choose when?

@do-me great question, wondering the same!