BAAI/bge-multilingual-gemma2

`BAAI/bge-multilingual-gemma2` powered by Text Generation Inference (TGI) - [Example Notebook](https://aka.ms/hugging-face-text-generation-streaming-online-endpoint) - [Original Model Card](https://h…

azureFoundry/HuggingFace/baai-bge-multilingual-gemma2@1

BAAI/bge-multilingual-gemma2 powered by Text Generation Inference (TGI)

Send Request

You can use cURL or any REST Client to send a request to the AzureML endpoint with your AzureML token.

curl <AZUREML_ENDPOINT_URL> \
    -X POST \
    -d '{"inputs":"What is Deep Learning?"}' \
    -H "Authorization: Bearer <AZUREML_TOKEN>" \
    -H "Content-Type: application/json"

Supported Parameters

  • inputs (string): Input prompt.
  • parameters (object):
    • best_of (integer): Generate best_of sequences and return the one if the highest token logprobs.
    • decoder_input_details (boolean): Whether to return decoder input token logprobs and ids.
    • details (boolean): Whether to return

Specifications & capabilities

Input
text
Output
embeddings
Published model ID
bge-multilingual-gemma2
Capabilities
embeddings

Evaluations

1–14 / 14

MTEB

87.976%

source id
mteb
leaderboard name
MTEB Classification
dimension
embedding
metric
Average task score
rank
55
sample size
12
aggregation scope
splits
  • test
subsets
  • default
  • en
  • fr
  • fra
  • pl
  • zh-CN
task_count
5
task_names
  • AmazonPolarityClassification
  • MTOPDomainClassification
  • MasakhaNEWSClassification
  • MassiveIntentClassification
  • MassiveScenarioClassification
sample_size
12
subset_count
6
subset_sample
  • default
  • en
  • fr
  • fra
  • pl
  • zh-CN
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

84.133%

source id
mteb
leaderboard name
MTEB Semantic Similarity
dimension
embedding
metric
Average task score
rank
77
sample size
7
aggregation scope
splits
  • test
subsets
  • default
  • fr
task_count
7
task_names
  • STS12
  • STS13
  • STS14
  • STS15
  • STS16
  • STSB
  • STSBenchmarkMultilingualSTS
sample_size
7
subset_count
2
subset_sample
  • default
  • fr
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

81.218%

source id
mteb
leaderboard name
MTEB Classification
dimension
embedding
metric
Average task score
rank
86
sample size
4
aggregation scope
splits
  • test
subsets
  • default
task_count
4
task_names
  • AmazonPolarityVNClassification
  • MTOPDomainVNClassification
  • MassiveIntentVNClassification
  • MassiveScenarioVNClassification
sample_size
4
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

67.605%

source id
mteb
leaderboard name
MTEB Other
dimension
embedding
metric
Average task score
rank
105
sample size
49
aggregation scope
splits
  • test
  • validation
subsets
  • default
  • fr
task_count
49
task_names
  • AFQMC
  • ATEC
  • AllegroReviews
  • AskUbuntuDupQuestions
  • BIOSSES
  • BQ
  • CBD
  • CDSC-E
  • CDSC-R
  • ClimateFEVER
  • Cmnli
  • DBPedia
  • DBPedia-PL
  • IFlyTek
  • JDReview
  • LCQMC
  • MultilingualSentiment
  • NFCorpus
  • NFCorpus-PL
  • Ocnli
  • OnlineShopping
  • OpusparcusPC
  • PAC
  • PAWSX
  • PSC
  • PolEmo2.0-IN
  • PolEmo2.0-OUT
  • PpcPC
  • QBQTC
  • Quora-PL
  • SCIDOCS
  • SCIDOCS-PL
  • SICK-E-PL
  • SICK-R
  • SICK-R-PL
  • SICKFr
  • SciDocsRR
  • SciFact
  • SciFact-PL
  • SprintDuplicateQuestions
  • SummEval
  • SummEvalFr
  • TNews
  • TRECCOVID
  • TRECCOVID-PL
  • Touche2020
  • TwitterSemEval2015
  • TwitterURLCorpus
  • Waimai
sample_size
49
subset_count
2
subset_sample
  • default
  • fr
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

56.772%

source id
mteb
leaderboard name
MTEB Clustering
dimension
embedding
metric
Average task score
rank
113
sample size
14
aggregation scope
splits
  • test
subsets
  • default
  • fr
  • fra
task_count
14
task_names
  • AlloProfClusteringP2P
  • AlloProfClusteringS2S
  • CLSClusteringP2P
  • CLSClusteringS2S
  • EightTagsClustering
  • HALClusteringS2S
  • MLSUMClusteringP2P
  • MLSUMClusteringS2S
  • MasakhaNEWSClusteringP2P
  • MasakhaNEWSClusteringS2S
  • StackExchangeClustering
  • StackExchangeClusteringP2P
  • ThuNewsClusteringP2P
  • ThuNewsClusteringS2S
sample_size
14
subset_count
3
subset_sample
  • default
  • fr
  • fra
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

68.911%

source id
mteb
leaderboard name
MTEB Overall
dimension
embedding
metric
Average task score
rank
119
sample size
97
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • default
  • en
  • fr
  • fra
  • fra-fra
  • pl
  • zh-CN
task_count
90
task_names
  • AFQMC
  • ATEC
  • AllegroReviews
  • AlloProfClusteringP2P
  • AlloProfClusteringS2S
  • AlloprofReranking
  • AlloprofRetrieval
  • AmazonPolarityClassification
  • AskUbuntuDupQuestions
  • BIOSSES
  • BQ
  • BSARDRetrieval
  • CBD
  • CDSC-E
  • CDSC-R
  • CLSClusteringP2P
  • CLSClusteringS2S
  • CQADupstackRetrieval
  • ClimateFEVER
  • Cmnli
  • CovidRetrieval
  • DBPedia
  • DBPedia-PL
  • EcomRetrieval
  • EightTagsClustering
  • HALClusteringS2S
  • IFlyTek
  • JDReview
  • LCQMC
  • MLSUMClusteringP2P
  • MLSUMClusteringS2S
  • MMarcoRetrieval
  • MTOPDomainClassification
  • MasakhaNEWSClassification
  • MasakhaNEWSClusteringP2P
  • MasakhaNEWSClusteringS2S
  • MassiveIntentClassification
  • MassiveScenarioClassification
  • MedicalRetrieval
  • MindSmallReranking
  • MintakaRetrieval
  • MultilingualSentiment
  • NFCorpus
  • NFCorpus-PL
  • Ocnli
  • OnlineShopping
  • OpusparcusPC
  • PAC
  • PAWSX
  • PSC
  • PolEmo2.0-IN
  • PolEmo2.0-OUT
  • PpcPC
  • QBQTC
  • Quora-PL
  • QuoraRetrieval
  • SCIDOCS
  • SCIDOCS-PL
  • SICK-E-PL
  • SICK-R
  • SICK-R-PL
  • SICKFr
  • STS12
  • STS13
  • STS14
  • STS15
  • STS16
  • STSB
  • STSBenchmarkMultilingualSTS
  • SciDocsRR
  • SciFact
  • SciFact-PL
  • SprintDuplicateQuestions
  • StackExchangeClustering
  • StackExchangeClusteringP2P
  • SummEval
  • SummEvalFr
  • SyntecReranking
  • SyntecRetrieval
  • TNews
  • TRECCOVID
  • TRECCOVID-PL
  • ThuNewsClusteringP2P
  • ThuNewsClusteringS2S
  • Touche2020
  • TwitterSemEval2015
  • TwitterURLCorpus
  • VideoRetrieval
  • Waimai
  • XPQARetrieval
sample_size
97
subset_count
7
subset_sample
  • default
  • en
  • fr
  • fra
  • fra-fra
  • pl
  • zh-CN
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

71.953%

source id
mteb
leaderboard name
MTEB Retrieval
dimension
embedding
metric
Average task score
rank
134
sample size
12
aggregation scope
splits
  • dev
  • test
subsets
  • default
  • fr
  • fra-fra
task_count
12
task_names
  • AlloprofRetrieval
  • BSARDRetrieval
  • CQADupstackRetrieval
  • CovidRetrieval
  • EcomRetrieval
  • MMarcoRetrieval
  • MedicalRetrieval
  • MintakaRetrieval
  • QuoraRetrieval
  • SyntecRetrieval
  • VideoRetrieval
  • XPQARetrieval
sample_size
12
subset_count
3
subset_sample
  • default
  • fr
  • fra-fra
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

67.416%

source id
mteb
leaderboard name
MTEB Reranking
dimension
embedding
metric
Average task score
rank
134
sample size
3
aggregation scope
splits
  • test
subsets
  • default
task_count
3
task_names
  • AlloprofReranking
  • MindSmallReranking
  • SyntecReranking
sample_size
3
subset_count
1
subset_sample
  • default
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • external
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

81.644%

source id
mteb
leaderboard name
MTEB Semantic Similarity
dimension
embedding
metric
Average task score
rank
140
sample size
13
aggregation scope
splits
  • test
  • validation
subsets
  • ar-ar
  • default
  • en-ar
  • en-de
  • en-en
  • en-tr
  • es-en
  • es-es
  • fr-en
  • it-en
  • ko-ko
  • nl-en
task_count
3
task_names
  • KLUE-STS
  • KorSTS
  • STS17
sample_size
13
subset_count
12
subset_sample
  • ar-ar
  • default
  • en-ar
  • en-de
  • en-en
  • en-tr
  • es-en
  • es-es
  • fr-en
  • it-en
  • ko-ko
  • nl-en
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

48.224%

source id
mteb
leaderboard name
MTEB Clustering
dimension
embedding
metric
Average task score
rank
211
sample size
203
aggregation scope
splits
  • test
subsets
  • ace_Latn
  • acm_Arab
  • acq_Arab
  • aeb_Arab
  • afr_Latn
  • ajp_Arab
  • aka_Latn
  • als_Latn
  • amh_Ethi
  • apc_Arab
  • arb_Latn
  • ars_Arab
  • ary_Arab
  • arz_Arab
  • asm_Beng
  • ast_Latn
  • awa_Deva
  • ayr_Latn
  • azb_Arab
  • azj_Latn
  • bak_Cyrl
  • bam_Latn
  • ban_Latn
  • bel_Cyrl
  • bem_Latn
  • ben_Beng
  • bho_Deva
  • bjn_Latn
  • bod_Tibt
  • bos_Latn
  • bug_Latn
  • bul_Cyrl
  • cat_Latn
  • ceb_Latn
  • ces_Latn
  • cjk_Latn
  • ckb_Arab
  • crh_Latn
  • cym_Latn
  • dan_Latn
  • default
  • deu_Latn
  • dik_Latn
  • dyu_Latn
  • dzo_Tibt
  • ell_Grek
  • eng_Latn
  • epo_Latn
  • est_Latn
  • eus_Latn
  • ewe_Latn
  • fao_Latn
  • fij_Latn
  • fin_Latn
  • fon_Latn
  • fra_Latn
  • fur_Latn
  • fuv_Latn
  • gaz_Latn
  • gla_Latn
  • gle_Latn
  • glg_Latn
  • grn_Latn
  • guj_Gujr
  • hat_Latn
  • hau_Latn
  • heb_Hebr
  • hin_Deva
  • hne_Deva
  • hrv_Latn
  • hun_Latn
  • hye_Armn
  • ibo_Latn
  • ilo_Latn
  • ind_Latn
  • isl_Latn
  • ita_Latn
  • jav_Latn
  • jpn_Jpan
  • kab_Latn
  • kac_Latn
  • kam_Latn
  • kan_Knda
  • kas_Deva
  • kat_Geor
  • kaz_Cyrl
  • kbp_Latn
  • kea_Latn
  • khk_Cyrl
  • khm_Khmr
  • kik_Latn
  • kin_Latn
  • kir_Cyrl
  • kmb_Latn
  • kmr_Latn
  • knc_Latn
  • kon_Latn
  • kor_Hang
  • lao_Laoo
  • lij_Latn
  • lim_Latn
  • lin_Latn
  • lit_Latn
  • lmo_Latn
  • ltg_Latn
  • ltz_Latn
  • lua_Latn
  • lug_Latn
  • luo_Latn
  • lus_Latn
  • lvs_Latn
  • mag_Deva
  • mai_Deva
  • mal_Mlym
  • mar_Deva
  • min_Latn
  • mkd_Cyrl
  • mlt_Latn
  • mni_Beng
  • mos_Latn
  • mri_Latn
  • mya_Mymr
  • nld_Latn
  • nno_Latn
  • nob_Latn
  • npi_Deva
  • nqo_Nkoo
  • nso_Latn
  • nus_Latn
  • nya_Latn
  • oci_Latn
  • ory_Orya
  • pag_Latn
  • pan_Guru
  • pap_Latn
  • pbt_Arab
  • pes_Arab
  • plt_Latn
  • pol_Latn
  • por_Latn
  • prs_Arab
  • quy_Latn
  • ron_Latn
  • run_Latn
  • rus_Cyrl
  • sag_Latn
  • san_Deva
  • sat_Olck
  • scn_Latn
  • shn_Mymr
  • sin_Sinh
  • slk_Latn
  • slv_Latn
  • smo_Latn
  • sna_Latn
  • snd_Arab
  • som_Latn
  • sot_Latn
  • spa_Latn
  • srd_Latn
  • srp_Cyrl
  • ssw_Latn
  • sun_Latn
  • swe_Latn
  • swh_Latn
  • szl_Latn
  • tam_Taml
  • taq_Tfng
  • tat_Cyrl
  • tel_Telu
  • tgk_Cyrl
  • tgl_Latn
  • tha_Thai
  • tir_Ethi
  • tpi_Latn
  • tsn_Latn
  • tso_Latn
  • tuk_Latn
  • tum_Latn
  • tur_Latn
  • twi_Latn
  • tzm_Tfng
  • uig_Arab
  • ukr_Cyrl
  • umb_Latn
  • urd_Arab
  • uzn_Latn
  • vec_Latn
  • vie_Latn
  • war_Latn
  • wol_Latn
  • xho_Latn
  • ydd_Hebr
  • yor_Latn
  • yue_Hant
  • zho_Hant
  • zsm_Latn
  • zul_Latn
task_count
7
task_names
  • KlueMrcDomainClustering
  • KlueYnatMrcCategoryClustering
  • PlscClusteringP2P
  • PlscClusteringS2S
  • SIB200ClusteringS2S
  • StackExchangeClustering-VN
  • StackExchangeClusteringP2P-VN
sample_size
203
subset_count
198
subset_sample
  • ace_Latn
  • acm_Arab
  • acq_Arab
  • aeb_Arab
  • afr_Latn
  • ajp_Arab
  • aka_Latn
  • als_Latn
  • amh_Ethi
  • apc_Arab
  • arb_Latn
  • ars_Arab
  • ary_Arab
  • arz_Arab
  • asm_Beng
  • ast_Latn
  • awa_Deva
  • ayr_Latn
  • azb_Arab
  • azj_Latn
  • bak_Cyrl
  • bam_Latn
  • ban_Latn
  • bel_Cyrl
  • bem_Latn
  • ben_Beng
  • bho_Deva
  • bjn_Latn
  • bod_Tibt
  • bos_Latn
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

57.434%

source id
mteb
leaderboard name
MTEB Pair Classification
dimension
embedding
metric
Average task score
rank
232
sample size
14
aggregation scope
splits
  • test
  • validation
subsets
  • de
  • en
  • es
  • fr
  • ja
  • ko
  • zh
task_count
1
task_names
  • PawsXPairClassification
sample_size
14
subset_count
7
subset_sample
  • de
  • en
  • es
  • fr
  • ja
  • ko
  • zh
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

61.183%

source id
mteb
leaderboard name
MTEB Retrieval
dimension
embedding
metric
Average task score
rank
311
sample size
8
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • default
  • eng_Latn-kor_Hang
  • ko
  • kor_Hang-eng_Latn
  • kor_Hang-kor_Hang
task_count
5
task_names
  • AutoRAGRetrieval
  • BelebeleRetrieval
  • MultiLongDocRetrieval
  • SQuADKorV1Retrieval
  • VieQuADRetrieval
sample_size
8
subset_count
5
subset_sample
  • default
  • eng_Latn-kor_Hang
  • ko
  • kor_Hang-eng_Latn
  • kor_Hang-kor_Hang
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

49.548%

source id
mteb
leaderboard name
MTEB Other
dimension
embedding
metric
Average task score
rank
496
sample size
225
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • Azure_azure-sdk-for-python
  • Character
  • ClusterHQ_flocker
  • Covid-QA
  • DataDog_integrations-core
  • ESG-Reports
  • LMEB_SciFact
  • LongDepQA
  • MLDR
  • Meaning
  • Microsoft_TypeScript
  • PeerQA
  • Plot
  • PokemonGoF_PokemonGo-Bot
  • QASPER
  • Qiskit_qiskit-terra
  • Relation
  • Setting
  • ShortDepQA
  • Span
  • StackStorm_st2
  • Times
  • ToolBench
  • abstention_evidence
  • adversarial
  • aggregative
  • apache_airflow
  • apache_incubator-airflow
  • appworld_qwen3_14b__generalized_query
  • appworld_qwen3_32b__generalized_query
  • appworld_qwen3_8b__generalized_query
  • assistant_facts_evidence
  • bfcl_qwen3_14b__generalized_query
  • bfcl_qwen3_14b__task_query
  • bfcl_qwen3_32b__generalized_query
  • bfcl_qwen3_32b__task_query
  • bfcl_qwen3_8b__generalized_query
  • bfcl_qwen3_8b__task_query
  • certbot_certbot
  • changing_evidence
  • commonsense
  • comparative
  • content_time_qs
  • dask_dask
  • datalad_datalad
  • date_span_time_qs
  • dates_time_qs
  • day_span_time_qs
  • default
  • default_claude_long__Entities
  • default_claude_long__Event_contents
  • default_claude_long__Full_event_details
  • default_claude_long__Other_entities
  • default_claude_long__Spaces
  • default_claude_long__Times
  • default_claude_short__Entities
  • default_claude_short__Event_contents
  • default_claude_short__Full_event_details
  • default_claude_short__Other_entities
  • default_claude_short__Spaces
  • default_claude_short__Times
  • default_claude_very_long__Entities
  • default_claude_very_long__Event_contents
  • default_claude_very_long__Full_event_details
  • default_claude_very_long__Other_entities
  • default_claude_very_long__Spaces
  • default_claude_very_long__Times
  • default_gpt4o_long__Entities
  • default_gpt4o_long__Event_contents
  • default_gpt4o_long__Full_event_details
  • default_gpt4o_long__Other_entities
  • default_gpt4o_long__Spaces
  • default_gpt4o_long__Times
  • default_gpt4o_short__Entities
  • default_gpt4o_short__Event_contents
  • default_gpt4o_short__Full_event_details
  • default_gpt4o_short__Other_entities
  • default_gpt4o_short__Spaces
  • default_gpt4o_short__Times
  • django_django
  • dmwm_WMCore
  • earlier_today_time_qs
  • easy
  • emotion
  • encode_django-rest-framework
  • event_driven__adversarial_abstention
  • event_driven__information_extraction
  • event_driven__mind-body_interaction
  • event_driven__mnestic_trigger_analysis
  • event_driven__temporal_reasoning
  • facebook_react
  • flashback_intensive__adversarial_abstention
  • flashback_intensive__information_extraction
  • flashback_intensive__mind-body_interaction
  • flashback_intensive__mnestic_trigger_analysis
  • flashback_intensive__temporal_reasoning
  • gorilla_huggingface
  • gorilla_pytorch
  • gorilla_tensor
  • great-expectations_great_expectations
  • hard
  • home-assistant_core
  • home-assistant_home-assistant
  • huggingface_transformers
  • implicit_connection_evidence
  • knowledge_update
  • knowledge_updating
  • korean
  • kubernetes_kubernetes
  • last_named_day_time_qs
  • medium
  • mesonbuild_meson
  • microsoft_vscode
  • mne-tools_mne-python
  • moby_moby
  • month_time_qs
  • multi_hop
  • multi_session
  • multi_session_assistant
  • napari_napari
  • numpy_numpy
  • open_domain
  • optuna_optuna
  • pandas-dev_pandas
  • post_processing
  • preference
  • preference_evidence
  • psychological_depth__adversarial_abstention
  • psychological_depth__information_extraction
  • psychological_depth__mind-body_interaction
  • psychological_depth__mnestic_trigger_analysis
  • psychological_depth__temporal_reasoning
  • pydata_xarray
  • pypa_pip
  • pytest-dev_pytest
  • pytorch_pytorch
  • raiden-network_raiden
  • rel_day_time_qs
  • rel_month_time_qs
  • rel_session_time_qs
  • rust-lang_rust
  • saltstack_salt
  • sci_fi_claude_long__Entities
  • sci_fi_claude_long__Event_contents
  • sci_fi_claude_long__Full_event_details
  • sci_fi_claude_long__Other_entities
  • sci_fi_claude_long__Spaces
  • sci_fi_claude_long__Times
  • sci_fi_claude_short__Entities
  • sci_fi_claude_short__Event_contents
  • sci_fi_claude_short__Full_event_details
  • sci_fi_claude_short__Other_entities
  • sci_fi_claude_short__Spaces
  • sci_fi_claude_short__Times
  • scikit-learn_scikit-learn
  • scipy_scipy
  • scrapy_scrapy
  • session_span_time_qs
  • session_time_qs
  • shopping_level1
  • shopping_level2
  • shopping_level3
  • single_hop
  • single_session_assistant
  • single_session_preference
  • single_session_user
  • spack_spack
  • sphinx-doc_sphinx
  • spotify_luigi
  • spring-projects_spring-framework
  • spyder-ide_spyder
  • sympy_sympy
  • temporal_reasoning
  • tensorflow_tensorflow
  • user_evidence
  • webpack_webpack
  • world_news_claude_long__Entities
  • world_news_claude_long__Event_contents
  • world_news_claude_long__Full_event_details
  • world_news_claude_long__Other_entities
  • world_news_claude_long__Spaces
  • world_news_claude_long__Times
  • world_news_claude_short__Entities
  • world_news_claude_short__Event_contents
  • world_news_claude_short__Full_event_details
  • world_news_claude_short__Other_entities
  • world_news_claude_short__Spaces
  • world_news_claude_short__Times
  • xonsh_xonsh
task_count
53
task_names
  • AskUbuntuDupQuestions-VN
  • BIOSSES-VN
  • CQADupstackAndroid-VN
  • CQADupstackGis-VN
  • CQADupstackMathematica-VN
  • CQADupstackPhysics-VN
  • CQADupstackProgrammers-VN
  • CQADupstackStats-VN
  • CQADupstackTex-VN
  • CQADupstackUnix-VN
  • CQADupstackWebmasters-VN
  • CQADupstackWordpress-VN
  • ClimateFEVER-VN
  • ConvoMem
  • CovidQA
  • DBPedia-VN
  • DeepPlanning
  • EPBench
  • ESGReports
  • Gorilla
  • KLUE-NLI
  • KLUE-TC
  • KnowMeBench
  • Ko-StrategyQA
  • KorNLI
  • LMEBMLDR
  • LMEB_SciFact
  • LawIRKo
  • LoCoMo
  • LongMemEval
  • LooGLE
  • MemBench
  • MemGovern
  • NFCorpus-VN
  • NovelQA
  • PeerQA
  • ProceduralMemBench
  • PublicHealthQA
  • QASPER
  • Quora-VN
  • REALTALK
  • ReMe
  • SCIDOCS-VN
  • SICK-R-VN
  • SciDocsRR-VN
  • SciFact-VN
  • SprintDuplicateQuestions-VN
  • TMD
  • TRECCOVID-VN
  • ToolBench
  • Touche2020-VN
  • TwitterSemEval2015-VN
  • TwitterURLCorpus-VN
sample_size
225
subset_count
189
subset_sample
  • Azure_azure-sdk-for-python
  • Character
  • ClusterHQ_flocker
  • Covid-QA
  • DataDog_integrations-core
  • ESG-Reports
  • LMEB_SciFact
  • LongDepQA
  • MLDR
  • Meaning
  • Microsoft_TypeScript
  • PeerQA
  • Plot
  • PokemonGoF_PokemonGo-Bot
  • QASPER
  • Qiskit_qiskit-terra
  • Relation
  • Setting
  • ShortDepQA
  • Span
  • StackStorm_st2
  • Times
  • ToolBench
  • abstention_evidence
  • adversarial
  • aggregative
  • apache_airflow
  • apache_incubator-airflow
  • appworld_qwen3_14b__generalized_query
  • appworld_qwen3_32b__generalized_query
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

MTEB

53.38%

source id
mteb
leaderboard name
MTEB Overall
dimension
embedding
metric
Average task score
rank
499
sample size
467
aggregation scope
splits
  • dev
  • test
  • validation
subsets
  • Azure_azure-sdk-for-python
  • Character
  • ClusterHQ_flocker
  • Covid-QA
  • DataDog_integrations-core
  • ESG-Reports
  • LMEB_SciFact
  • LongDepQA
  • MLDR
  • Meaning
  • Microsoft_TypeScript
  • PeerQA
  • Plot
  • PokemonGoF_PokemonGo-Bot
  • QASPER
  • Qiskit_qiskit-terra
  • Relation
  • Setting
  • ShortDepQA
  • Span
  • StackStorm_st2
  • Times
  • ToolBench
  • abstention_evidence
  • ace_Latn
  • acm_Arab
  • acq_Arab
  • adversarial
  • aeb_Arab
  • afr_Latn
  • aggregative
  • ajp_Arab
  • aka_Latn
  • als_Latn
  • amh_Ethi
  • apache_airflow
  • apache_incubator-airflow
  • apc_Arab
  • appworld_qwen3_14b__generalized_query
  • appworld_qwen3_32b__generalized_query
  • appworld_qwen3_8b__generalized_query
  • ar-ar
  • arb_Latn
  • ars_Arab
  • ary_Arab
  • arz_Arab
  • asm_Beng
  • assistant_facts_evidence
  • ast_Latn
  • awa_Deva
  • ayr_Latn
  • azb_Arab
  • azj_Latn
  • bak_Cyrl
  • bam_Latn
  • ban_Latn
  • bel_Cyrl
  • bem_Latn
  • ben_Beng
  • bfcl_qwen3_14b__generalized_query
  • bfcl_qwen3_14b__task_query
  • bfcl_qwen3_32b__generalized_query
  • bfcl_qwen3_32b__task_query
  • bfcl_qwen3_8b__generalized_query
  • bfcl_qwen3_8b__task_query
  • bho_Deva
  • bjn_Latn
  • bod_Tibt
  • bos_Latn
  • bug_Latn
  • bul_Cyrl
  • cat_Latn
  • ceb_Latn
  • certbot_certbot
  • ces_Latn
  • changing_evidence
  • cjk_Latn
  • ckb_Arab
  • commonsense
  • comparative
  • content_time_qs
  • crh_Latn
  • cym_Latn
  • dan_Latn
  • dask_dask
  • datalad_datalad
  • date_span_time_qs
  • dates_time_qs
  • day_span_time_qs
  • de
  • default
  • default_claude_long__Entities
  • default_claude_long__Event_contents
  • default_claude_long__Full_event_details
  • default_claude_long__Other_entities
  • default_claude_long__Spaces
  • default_claude_long__Times
  • default_claude_short__Entities
  • default_claude_short__Event_contents
  • default_claude_short__Full_event_details
  • default_claude_short__Other_entities
  • default_claude_short__Spaces
  • default_claude_short__Times
  • default_claude_very_long__Entities
  • default_claude_very_long__Event_contents
  • default_claude_very_long__Full_event_details
  • default_claude_very_long__Other_entities
  • default_claude_very_long__Spaces
  • default_claude_very_long__Times
  • default_gpt4o_long__Entities
  • default_gpt4o_long__Event_contents
  • default_gpt4o_long__Full_event_details
  • default_gpt4o_long__Other_entities
  • default_gpt4o_long__Spaces
  • default_gpt4o_long__Times
  • default_gpt4o_short__Entities
  • default_gpt4o_short__Event_contents
  • default_gpt4o_short__Full_event_details
  • default_gpt4o_short__Other_entities
  • default_gpt4o_short__Spaces
  • default_gpt4o_short__Times
  • deu_Latn
  • dik_Latn
  • django_django
  • dmwm_WMCore
  • dyu_Latn
  • dzo_Tibt
  • earlier_today_time_qs
  • easy
  • ell_Grek
  • emotion
  • en
  • en-ar
  • en-de
  • en-en
  • en-tr
  • encode_django-rest-framework
  • eng_Latn
  • eng_Latn-kor_Hang
  • epo_Latn
  • es
  • es-en
  • es-es
  • est_Latn
  • eus_Latn
  • event_driven__adversarial_abstention
  • event_driven__information_extraction
  • event_driven__mind-body_interaction
  • event_driven__mnestic_trigger_analysis
  • event_driven__temporal_reasoning
  • ewe_Latn
  • facebook_react
  • fao_Latn
  • fij_Latn
  • fin_Latn
  • flashback_intensive__adversarial_abstention
  • flashback_intensive__information_extraction
  • flashback_intensive__mind-body_interaction
  • flashback_intensive__mnestic_trigger_analysis
  • flashback_intensive__temporal_reasoning
  • fon_Latn
  • fr
  • fr-en
  • fra_Latn
  • fur_Latn
  • fuv_Latn
  • gaz_Latn
  • gla_Latn
  • gle_Latn
  • glg_Latn
  • gorilla_huggingface
  • gorilla_pytorch
  • gorilla_tensor
  • great-expectations_great_expectations
  • grn_Latn
  • guj_Gujr
  • hard
  • hat_Latn
  • hau_Latn
  • heb_Hebr
  • hin_Deva
  • hne_Deva
  • home-assistant_core
  • home-assistant_home-assistant
  • hrv_Latn
  • huggingface_transformers
  • hun_Latn
  • hye_Armn
  • ibo_Latn
  • ilo_Latn
  • implicit_connection_evidence
  • ind_Latn
  • isl_Latn
  • it-en
  • ita_Latn
  • ja
  • jav_Latn
  • jpn_Jpan
  • kab_Latn
  • kac_Latn
  • kam_Latn
  • kan_Knda
  • kas_Deva
  • kat_Geor
  • kaz_Cyrl
  • kbp_Latn
  • kea_Latn
  • khk_Cyrl
  • khm_Khmr
  • kik_Latn
  • kin_Latn
  • kir_Cyrl
  • kmb_Latn
  • kmr_Latn
  • knc_Latn
  • knowledge_update
  • knowledge_updating
  • ko
  • ko-ko
  • kon_Latn
  • kor_Hang
  • kor_Hang-eng_Latn
  • kor_Hang-kor_Hang
  • korean
  • kubernetes_kubernetes
  • lao_Laoo
  • last_named_day_time_qs
  • lij_Latn
  • lim_Latn
  • lin_Latn
  • lit_Latn
  • lmo_Latn
  • ltg_Latn
  • ltz_Latn
  • lua_Latn
  • lug_Latn
  • luo_Latn
  • lus_Latn
  • lvs_Latn
  • mag_Deva
  • mai_Deva
  • mal_Mlym
  • mar_Deva
  • medium
  • mesonbuild_meson
  • microsoft_vscode
  • min_Latn
  • mkd_Cyrl
  • mlt_Latn
  • mne-tools_mne-python
  • mni_Beng
  • moby_moby
  • month_time_qs
  • mos_Latn
  • mri_Latn
  • multi_hop
  • multi_session
  • multi_session_assistant
  • mya_Mymr
  • napari_napari
  • nl-en
  • nld_Latn
  • nno_Latn
  • nob_Latn
  • npi_Deva
  • nqo_Nkoo
  • nso_Latn
  • numpy_numpy
  • nus_Latn
  • nya_Latn
  • oci_Latn
  • open_domain
  • optuna_optuna
  • ory_Orya
  • pag_Latn
  • pan_Guru
  • pandas-dev_pandas
  • pap_Latn
  • pbt_Arab
  • pes_Arab
  • plt_Latn
  • pol_Latn
  • por_Latn
  • post_processing
  • preference
  • preference_evidence
  • prs_Arab
  • psychological_depth__adversarial_abstention
  • psychological_depth__information_extraction
  • psychological_depth__mind-body_interaction
  • psychological_depth__mnestic_trigger_analysis
  • psychological_depth__temporal_reasoning
  • pydata_xarray
  • pypa_pip
  • pytest-dev_pytest
  • pytorch_pytorch
  • quy_Latn
  • raiden-network_raiden
  • rel_day_time_qs
  • rel_month_time_qs
  • rel_session_time_qs
  • ron_Latn
  • run_Latn
  • rus_Cyrl
  • rust-lang_rust
  • sag_Latn
  • saltstack_salt
  • san_Deva
  • sat_Olck
  • sci_fi_claude_long__Entities
  • sci_fi_claude_long__Event_contents
  • sci_fi_claude_long__Full_event_details
  • sci_fi_claude_long__Other_entities
  • sci_fi_claude_long__Spaces
  • sci_fi_claude_long__Times
  • sci_fi_claude_short__Entities
  • sci_fi_claude_short__Event_contents
  • sci_fi_claude_short__Full_event_details
  • sci_fi_claude_short__Other_entities
  • sci_fi_claude_short__Spaces
  • sci_fi_claude_short__Times
  • scikit-learn_scikit-learn
  • scipy_scipy
  • scn_Latn
  • scrapy_scrapy
  • session_span_time_qs
  • session_time_qs
  • shn_Mymr
  • shopping_level1
  • shopping_level2
  • shopping_level3
  • sin_Sinh
  • single_hop
  • single_session_assistant
  • single_session_preference
  • single_session_user
  • slk_Latn
  • slv_Latn
  • smo_Latn
  • sna_Latn
  • snd_Arab
  • som_Latn
  • sot_Latn
  • spa_Latn
  • spack_spack
  • sphinx-doc_sphinx
  • spotify_luigi
  • spring-projects_spring-framework
  • spyder-ide_spyder
  • srd_Latn
  • srp_Cyrl
  • ssw_Latn
  • sun_Latn
  • swe_Latn
  • swh_Latn
  • sympy_sympy
  • szl_Latn
  • tam_Taml
  • taq_Tfng
  • tat_Cyrl
  • tel_Telu
  • temporal_reasoning
  • tensorflow_tensorflow
  • tgk_Cyrl
  • tgl_Latn
  • tha_Thai
  • tir_Ethi
  • tpi_Latn
  • tsn_Latn
  • tso_Latn
  • tuk_Latn
  • tum_Latn
  • tur_Latn
  • twi_Latn
  • tzm_Tfng
  • uig_Arab
  • ukr_Cyrl
  • umb_Latn
  • urd_Arab
  • user_evidence
  • uzn_Latn
  • vec_Latn
  • vie_Latn
  • war_Latn
  • webpack_webpack
  • wol_Latn
  • world_news_claude_long__Entities
  • world_news_claude_long__Event_contents
  • world_news_claude_long__Full_event_details
  • world_news_claude_long__Other_entities
  • world_news_claude_long__Spaces
  • world_news_claude_long__Times
  • world_news_claude_short__Entities
  • world_news_claude_short__Event_contents
  • world_news_claude_short__Full_event_details
  • world_news_claude_short__Other_entities
  • world_news_claude_short__Spaces
  • world_news_claude_short__Times
  • xho_Latn
  • xonsh_xonsh
  • ydd_Hebr
  • yor_Latn
  • yue_Hant
  • zh
  • zho_Hant
  • zsm_Latn
  • zul_Latn
task_count
73
task_names
  • AmazonPolarityVNClassification
  • AskUbuntuDupQuestions-VN
  • AutoRAGRetrieval
  • BIOSSES-VN
  • BelebeleRetrieval
  • CQADupstackAndroid-VN
  • CQADupstackGis-VN
  • CQADupstackMathematica-VN
  • CQADupstackPhysics-VN
  • CQADupstackProgrammers-VN
  • CQADupstackStats-VN
  • CQADupstackTex-VN
  • CQADupstackUnix-VN
  • CQADupstackWebmasters-VN
  • CQADupstackWordpress-VN
  • ClimateFEVER-VN
  • ConvoMem
  • CovidQA
  • DBPedia-VN
  • DeepPlanning
  • EPBench
  • ESGReports
  • Gorilla
  • KLUE-NLI
  • KLUE-STS
  • KLUE-TC
  • KlueMrcDomainClustering
  • KlueYnatMrcCategoryClustering
  • KnowMeBench
  • Ko-StrategyQA
  • KorNLI
  • KorSTS
  • LMEBMLDR
  • LMEB_SciFact
  • LawIRKo
  • LoCoMo
  • LongMemEval
  • LooGLE
  • MTOPDomainVNClassification
  • MassiveIntentVNClassification
  • MassiveScenarioVNClassification
  • MemBench
  • MemGovern
  • MultiLongDocRetrieval
  • NFCorpus-VN
  • NovelQA
  • PawsXPairClassification
  • PeerQA
  • PlscClusteringP2P
  • PlscClusteringS2S
  • ProceduralMemBench
  • PublicHealthQA
  • QASPER
  • Quora-VN
  • REALTALK
  • ReMe
  • SCIDOCS-VN
  • SIB200ClusteringS2S
  • SICK-R-VN
  • SQuADKorV1Retrieval
  • STS17
  • SciDocsRR-VN
  • SciFact-VN
  • SprintDuplicateQuestions-VN
  • StackExchangeClustering-VN
  • StackExchangeClusteringP2P-VN
  • TMD
  • TRECCOVID-VN
  • ToolBench
  • Touche2020-VN
  • TwitterSemEval2015-VN
  • TwitterURLCorpus-VN
  • VieQuADRetrieval
sample_size
467
subset_count
407
subset_sample
  • Azure_azure-sdk-for-python
  • Character
  • ClusterHQ_flocker
  • Covid-QA
  • DataDog_integrations-core
  • ESG-Reports
  • LMEB_SciFact
  • LongDepQA
  • MLDR
  • Meaning
  • Microsoft_TypeScript
  • PeerQA
  • Plot
  • PokemonGoF_PokemonGo-Bot
  • QASPER
  • Qiskit_qiskit-terra
  • Relation
  • Setting
  • ShortDepQA
  • Span
  • StackStorm_st2
  • Times
  • ToolBench
  • abstention_evidence
  • ace_Latn
  • acm_Arab
  • acq_Arab
  • adversarial
  • aeb_Arab
  • afr_Latn
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • 992e13d8984fde2c31ef8a3cb2c038aeec513b8a
source model id
BAAI/bge-multilingual-gemma2
Original result ↗

Provider specifications & access

1–2 / 2

BAAI/bge-multilingual-gemma2 ↗

source model id
HuggingFace/baai-bge-multilingual-gemma2@1
hosting
foundry-managed-compute
platforms
  • Microsoft Foundry
license
gemma
architecture
input_modalities
  • text
output_modalities
  • embeddings
capabilities

    BAAI/bge-multilingual-gemma2 ↗

    source model id
    HuggingFace/baai-bge-multilingual-gemma2@2
    hosting
    foundry-managed-compute
    platforms
    • Microsoft Foundry
    license
    gemma
    architecture
    input_modalities
    • text
    output_modalities
    • embeddings
    capabilities

      Original sources

      1–6 / 6

      Hugging Face · BAAI/bge-multilingual-gemma2 ↗

      Microsoft Foundry · Model page ↗

      Microsoft Foundry · Model page ↗

      Microsoft Foundry · Model documentation ↗

      Hugging Face · Catalog ↗

      Microsoft Foundry · Catalog ↗

      Source links

      1–2 / 2

      Explore more models

      Models

      Loading
      EVIDENCE / COMPARISON

      Benchmarks

      Explore every benchmark

      Loading

      These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

      Install TensorTrace