dragonkue / BGE-m3-ko

huggingface.co
Total runs: 76.5K
24-hour runs: -3.1K
7-day runs: -6.7K
30-day runs: -183.8K
Model's Last Updated: October 16 2025
sentence-similarity

Introduction of BGE-m3-ko

Model Details of BGE-m3-ko

SentenceTransformer

This is a sentence-transformers model trained on the train_set dataset. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details
  • Learning other languages โ€‹โ€‹besides Chinese and English is insufficient, so additional learning is needed to optimize use of other languages.
  • This model is additionally trained on the Korean dataset.
Model Description
  • Model Type: Sentence Transformer Transformer Encoder
  • Maximum Sequence Length: 8192 tokens
  • Output Dimensionality: 1024 tokens
  • Similarity Function: Cosine Similarity
  • Training Dataset:
Model Sources
Full Model Architecture
SentenceTransformer(
  (0): Transformer({'max_seq_length': 8192, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)
Usage
Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the ๐Ÿค— Hub
model = SentenceTransformer("dragonkue/bge-m3-ko")
# Run inference
sentences = [
    '์ˆ˜๊ธ‰๊ถŒ์ž ์ค‘ ๊ทผ๋กœ ๋Šฅ๋ ฅ์ด ์—†๋Š” ์ž„์‚ฐ๋ถ€๋Š” ๋ช‡ ์ข…์— ํ•ด๋‹นํ•˜๋‹ˆ?',
    '๋‚ด๋…„๋ถ€ํ„ฐ ์ €์†Œ๋“์ธต 1์„ธ ๋ฏธ๋งŒ ์•„๋™์˜ \n์˜๋ฃŒ๋น„ ๋ถ€๋‹ด์ด ๋” ๋‚ฎ์•„์ง„๋‹ค!\n์˜๋ฃŒ๊ธ‰์—ฌ์ œ๋„ ๊ฐœ์š”\nโ–ก (๋ชฉ์ ) ์ƒํ™œ์œ ์ง€ ๋Šฅ๋ ฅ์ด ์—†๊ฑฐ๋‚˜ ์ƒํ™œ์ด ์–ด๋ ค์šด ๊ตญ๋ฏผ๋“ค์—๊ฒŒ ๋ฐœ์ƒํ•˜๋Š” ์งˆ๋ณ‘, ๋ถ€์ƒ, ์ถœ์‚ฐ ๋“ฑ์— ๋Œ€ํ•ด ๊ตญ๊ฐ€๊ฐ€ ์˜๋ฃŒ์„œ๋น„์Šค ์ œ๊ณต\nโ–ก (์ง€์›๋Œ€์ƒ) ๊ตญ๋ฏผ๊ธฐ์ดˆ์ƒํ™œ๋ณด์žฅ ์ˆ˜๊ธ‰๊ถŒ์ž, ํƒ€ ๋ฒ•์— ์˜ํ•œ ์ˆ˜๊ธ‰๊ถŒ์ž ๋“ฑ\n\n| ๊ตฌ๋ถ„ | ๊ตญ๋ฏผ๊ธฐ์ดˆ์ƒํ™œ๋ณด์žฅ๋ฒ•์— ์˜ํ•œ ์ˆ˜๊ธ‰๊ถŒ์ž | ๊ตญ๋ฏผ๊ธฐ์ดˆ์ƒํ™œ๋ณด์žฅ๋ฒ• ์ด์™ธ์˜ ํƒ€ ๋ฒ•์— ์˜ํ•œ ์ˆ˜๊ธ‰๊ถŒ์ž |\n| --- | --- | --- |\n| 1์ข… | โ—‹ ๊ตญ๋ฏผ๊ธฐ์ดˆ์ƒํ™œ๋ณด์žฅ ์ˆ˜๊ธ‰๊ถŒ์ž ์ค‘ ๊ทผ๋กœ๋Šฅ๋ ฅ์ด ์—†๋Š” ์ž๋งŒ์œผ๋กœ ๊ตฌ์„ฑ๋œ ๊ฐ€๊ตฌ - 18์„ธ ๋ฏธ๋งŒ, 65์„ธ ์ด์ƒ - 4๊ธ‰ ์ด๋‚ด ์žฅ์• ์ธ - ์ž„์‚ฐ๋ถ€, ๋ณ‘์—ญ์˜๋ฌด์ดํ–‰์ž ๋“ฑ | โ—‹ ์ด์žฌ๋ฏผ(์žฌํ•ด๊ตฌํ˜ธ๋ฒ•) โ—‹ ์˜์ƒ์ž ๋ฐ ์˜์‚ฌ์ž์˜ ์œ ์กฑโ—‹ ๊ตญ๋‚ด ์ž…์–‘๋œ 18์„ธ ๋ฏธ๋งŒ ์•„๋™โ—‹ ๊ตญ๊ฐ€์œ ๊ณต์ž ๋ฐ ๊ทธ ์œ ์กฑโ€ค๊ฐ€์กฑโ—‹ ๊ตญ๊ฐ€๋ฌดํ˜•๋ฌธํ™”์žฌ ๋ณด์œ ์ž ๋ฐ ๊ทธ ๊ฐ€์กฑโ—‹ ์ƒˆํ„ฐ๋ฏผ(๋ถํ•œ์ดํƒˆ์ฃผ๋ฏผ)๊ณผ ๊ทธ ๊ฐ€์กฑโ—‹ 5โ€ค18 ๋ฏผ์ฃผํ™”์šด๋™ ๊ด€๋ จ์ž ๋ฐ ๊ทธ ์œ ๊ฐ€์กฑโ—‹ ๋…ธ์ˆ™์ธ โ€ป ํ–‰๋ คํ™˜์ž (์˜๋ฃŒ๊ธ‰์—ฌ๋ฒ• ์‹œํ–‰๋ น) |\n| 2์ข… | โ—‹ ๊ตญ๋ฏผ๊ธฐ์ดˆ์ƒํ™œ๋ณด์žฅ ์ˆ˜๊ธ‰๊ถŒ์ž ์ค‘ ๊ทผ๋กœ๋Šฅ๋ ฅ์ด ์žˆ๋Š” ๊ฐ€๊ตฌ | - |\n',
    '๋ฌธ์žฌ์ธ ๋Œ€ํ†ต๋ น์˜ ๊ตญ์ •์ˆ˜ํ–‰ ์ง€์ง€์œจ๊ณผ ๋”๋ถˆ์–ด๋ฏผ์ฃผ๋‹น์˜ ์ง€์ง€์œจ์ด 2์ฃผ ์—ฐ์† ์ƒ์Šนํ–ˆ๋‹ค๋Š” ์—ฌ๋ก ์กฐ์‚ฌ ๊ฒฐ๊ณผ๊ฐ€ ๋‚˜์™”๋‹ค. ํ•œ๋ฏธ์ •์ƒํšŒ๋‹ด์˜ ํšจ๊ณผ๋กœ ํ’€์ด๋œ๋‹ค. \nํ•œ๊ตญ๊ฐค๋Ÿฝ์€ 25~27์ผ ์ „๊ตญ ๋งŒ 18์„ธ ์ด์ƒ 1,003๋ช…์—๊ฒŒ ๋ฌธ ๋Œ€ํ†ต๋ น์˜ ์ง๋ฌด์ˆ˜ํ–‰ ํ‰๊ฐ€๋ฅผ ์กฐ์‚ฌํ•œ ๊ฒฐ๊ณผ(ํ‘œ๋ณธ์˜ค์ฐจ 95% ์‹ ๋ขฐ์ˆ˜์ค€์— ยฑ3.1%ํฌ์ธํŠธ), 37%๊ฐ€ ๊ธ์ • ํ‰๊ฐ€ํ–ˆ๋‹ค๊ณ  28์ผ ๋ฐํ˜”๋‹ค. \n๊ธ์ • ํ‰๊ฐ€๋Š” ์ง€๋‚œ์ฃผ๋ณด๋‹ค 3%ํฌ์ธํŠธ ์˜ฌ๋ž๋‹ค. ๋ถ€์ • ํ‰๊ฐ€๋Š” 52%๋กœ ์ง€๋‚œ์ฃผ๋ณด๋‹ค 6%ํฌ์ธํŠธ ๋–จ์–ด์กŒ๋‹ค. ์ตœ๊ทผ 60%๋Œ€๋ฅผ ๋„˜๋‚˜๋“ค๋˜ ๋ถ€์ • ํ‰๊ฐ€๊ฐ€ 50%๋Œ€ ์ดˆ๋ฐ˜์œผ๋กœ ํ•˜๋ฝํ–ˆ๋‹ค. 10%๋Š” ์˜๊ฒฌ์„ ์œ ๋ณดํ–ˆ๋‹ค. \n์ง€์—ญ๋ณ„๋กœ ๋ณด๋ฉด ์„œ์šธ๊ณผ ์ธ์ฒœยท๊ฒฝ๊ธฐ์˜ ๊ธ์ • ํ‰๊ฐ€๊ฐ€ ๊ฐ๊ฐ 37%๋ž€ ์ ์ด ๋ˆˆ์— ๋ˆ๋‹ค. ๋ถ€์‚ฐยท์šธ์‚ฐยท๊ฒฝ๋‚จ๋„ 33%๋กœ, ๋Œ€๊ตฌยท๊ฒฝ๋ถ(25%)๊ณผ ๋‹ฌ๋ฆฌ 30%๋Œ€๋กœ ๋‚˜ํƒ€๋‚ฌ๋‹ค. ์—ฐ๋ น๋ณ„๋กœ๋Š” 40๋Œ€๊ฐ€ 49%๋กœ ๊ฐ€์žฅ ๋†’์•˜๊ณ , 18~29์„ธ๋Š” 31%๋กœ ์ง‘๊ณ„๋๋‹ค. ์ •์น˜์  ์ด๋…ยท์„ฑํ–ฅ์ด ์ค‘๋„๋ผ๊ณ  ํ•œ ์‘๋‹ต์ž์˜ 34%๋Š” ๊ธ์ • ํ‰๊ฐ€ํ–ˆ๋‹ค. \nโ—‡๊ธ์ • ํ‰๊ฐ€ ์ด์œ  โ€˜์™ธ๊ตยท๊ตญ์ œ ๊ด€๊ณ„โ€™ 26%P ์ƒ์Šน \n๋ฌธ ๋Œ€ํ†ต๋ น์˜ ์ง€์ง€์œจ ์ƒ์Šน์€ ํ•œ๋ฏธ์ •์ƒํšŒ๋‹ด ์„ฑ๊ณผ๊ฐ€ ์˜ํ–ฅ์„ ๋ฏธ์นœ ๊ฒƒ์œผ๋กœ ๋ณด์ธ๋‹ค. ๊ธ์ • ํ‰๊ฐ€ ์ด์œ ๋กœ๋Š” โ€˜์™ธ๊ตยท๊ตญ์ œ ๊ด€๊ณ„โ€™๊ฐ€ ๊ฐ€์žฅ ๋†’์•˜๋‹ค. 30%๋กœ ์ง€๋‚œ์ฃผ๋ณด๋‹ค 26%ํฌ์ธํŠธ๋‚˜ ์˜ฌ๋ž๋‹ค. 15๊ฐœ์›”๊ฐ„ โ€˜์‹ ์ข… ์ฝ”๋กœ๋‚˜๋ฐ”์ด๋Ÿฌ์Šค ๊ฐ์—ผ์ฆ(์ฝ”๋กœ๋‚˜19) ๋Œ€์ฒ˜โ€™๊ฐ€ 1์œ„์˜€๋Š”๋ฐ, ํ•œ๋ฏธ์ •์ƒํšŒ๋‹ด ์ดํ›„ ์™ธ๊ตยท๊ตญ์ œ ๊ด€๊ณ„๋กœ ์ˆœ์œ„๊ฐ€ ๋ฐ”๋€Œ์—ˆ๋‹ค. \n๋‹ค์Œ์œผ๋กœ โ€˜์ฝ”๋กœ๋‚˜19 ๋Œ€์ฒ˜โ€™ 22%, โ€˜์ตœ์„ ์„ ๋‹คํ•จยท์—ด์‹ฌํžˆ ํ•œ๋‹คโ€™ 6%, โ€˜๋ถํ•œ ๊ด€๊ณ„โ€™ 4%, โ€˜์ „๋ฐ˜์ ์œผ๋กœ ์ž˜ํ•œ๋‹คโ€™ 4% ์ˆœ์ด์—ˆ๋‹ค. \n๋ถ€์ • ํ‰๊ฐ€ ์ด์œ ๋กœ๋Š” โ€˜๋ถ€๋™์‚ฐ ์ •์ฑ…โ€™์ด 29%๋กœ ๊ฐ€์žฅ ๋†’์•˜๋‹ค. ๋‹ค์Œ์œผ๋กœ โ€˜๊ฒฝ์ œยท๋ฏผ์ƒ ๋ฌธ์ œ ํ•ด๊ฒฐ ๋ถ€์กฑโ€™ 10%, โ€˜์ฝ”๋กœ๋‚˜19 ๋Œ€์ฒ˜ ๋ฏธํกโ€™ 5%, โ€˜๊ณต์ •ํ•˜์ง€ ๋ชปํ•จยท๋‚ด๋กœ๋‚จ๋ถˆโ€™ 5%, โ€˜์ธ์‚ฌ ๋ฌธ์ œโ€™ 4% ์ˆœ์ด์—ˆ๋‹ค. \n๋ฏผ์ฃผ๋‹น ์ง€์ง€์œจ ์—ญ์‹œ ๋ฌธ ๋Œ€ํ†ต๋ น ์ง€์ง€์œจ๊ณผ ๋งˆ์ฐฌ๊ฐ€์ง€๋กœ 2์ฃผ ์—ฐ์† ์ƒ์Šนํ–ˆ๋‹ค. ์ •๋‹น ์ง€์ง€๋„ ์กฐ์‚ฌ์—์„œ ๋ฏผ์ฃผ๋‹น์€ 34%๋กœ ์ง€๋‚œ์ฃผ๋ณด๋‹ค 2%ํฌ์ธํŠธ ์˜ฌ๋ž๋‹ค. \n๊ตญ๋ฏผ์˜ํž˜์€ 27%๋กœ ์ง€๋‚œ์ฃผ๋ณด๋‹ค 1%ํฌ์ธํŠธ ์˜ฌ๋ž๋‹ค. ๋ฏผ์ฃผ๋‹น๊ณผ ๊ตญ๋ฏผ์˜ํž˜์˜ ์ง€์ง€์œจ ๊ฒฉ์ฐจ๋Š” 7%ํฌ์ธํŠธ๋กœ, ์˜ค์ฐจ๋ฒ”์œ„ ๋ฐ–์„ ๋ฒ—์–ด๋‚ฌ๋‹ค. ๋‹ค์Œ์œผ๋กœ ์ •์˜๋‹น 5%, ์—ด๋ฆฐ๋ฏผ์ฃผ๋‹น 3%, ๊ตญ๋ฏผ์˜๋‹น 3% ์ˆœ์ด์—ˆ๋‹ค. ๋ฌด๋‹น์ธต์€ 27%๋กœ ์กฐ์‚ฌ๋๋‹ค. \nโ€ป์ž์„ธํ•œ ๋‚ด์šฉ์€ ํ•œ๊ตญ๊ฐค๋Ÿฝ ๋˜๋Š” ์ค‘์•™์„ ๊ฑฐ์—ฌ๋ก ์กฐ์‚ฌ์‹ฌ์˜์œ„์›ํšŒ ํ™ˆํŽ˜์ด์ง€๋ฅผ ์ฐธ์กฐํ•˜๋ฉด ๋œ๋‹ค.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]
Evaluation
Metrics
  • ndcg, mrr, map metrics are metrics that consider ranking, while accuracy, precision, and recall are metrics that do not consider ranking. (Example: When considering ranking for retrieval top 10, different scores are given when the correct document is in 1st place and when it is in 10th place. However, accuracy, precision, and recall scores are the same if they are in the top 10.)
Information Retrieval
  • Korean Embedding Benchmark is a benchmark with a relatively long 3/4 quantile of string length of 1024
Korean Embedding Benchmark with AutoRAG

This is a benchmark of Korean embedding models. ( https://github.com/Marker-Inc-Korea/AutoRAG-example-korean-embedding-benchmark )

  • Top-k 1
Model name F1 Recall Precision mAP mRR NDCG
paraphrase-multilingual-mpnet-base-v2 0.3596 0.3596 0.3596 0.3596 0.3596 0.3596
KoSimCSE-roberta 0.4298 0.4298 0.4298 0.4298 0.4298 0.4298
Cohere embed-multilingual-v3.0 0.3596 0.3596 0.3596 0.3596 0.3596 0.3596
openai ada 002 0.4737 0.4737 0.4737 0.4737 0.4737 0.4737
multilingual-e5-large-instruct 0.4649 0.4649 0.4649 0.4649 0.4649 0.4649
Upstage Embedding 0.6579 0.6579 0.6579 0.6579 0.6579 0.6579
paraphrase-multilingual-MiniLM-L12-v2 0.2982 0.2982 0.2982 0.2982 0.2982 0.2982
openai_embed_3_small 0.5439 0.5439 0.5439 0.5439 0.5439 0.5439
ko-sroberta-multitask 0.4211 0.4211 0.4211 0.4211 0.4211 0.4211
openai_embed_3_large 0.6053 0.6053 0.6053 0.6053 0.6053 0.6053
KU-HIAI-ONTHEIT-large-v1 0.7105 0.7105 0.7105 0.7105 0.7105 0.7105
KU-HIAI-ONTHEIT-large-v1.1 0.7193 0.7193 0.7193 0.7193 0.7193 0.7193
kf-deberta-multitask 0.4561 0.4561 0.4561 0.4561 0.4561 0.4561
gte-multilingual-base 0.5877 0.5877 0.5877 0.5877 0.5877 0.5877
BGE-m3 0.6578 0.6578 0.6578 0.6578 0.6578 0.6578
BGE-m3-ko 0.7456 0.7456 0.7456 0.7456 0.7456 0.7456
  • Top-k 3
Model name F1 Recall Precision mAP mRR NDCG
paraphrase-multilingual-mpnet-base-v2 0.2368 0.4737 0.1579 0.2032 0.2032 0.2712
KoSimCSE-roberta 0.3026 0.6053 0.2018 0.2661 0.2661 0.3515
Cohere embed-multilingual-v3.0 0.2851 0.5702 0.1901 0.2515 0.2515 0.3321
openai ada 002 0.3553 0.7105 0.2368 0.3202 0.3202 0.4186
multilingual-e5-large-instruct 0.3333 0.6667 0.2222 0.2909 0.2909 0.3856
Upstage Embedding 0.4211 0.8421 0.2807 0.3509 0.3509 0.4743
paraphrase-multilingual-MiniLM-L12-v2 0.2061 0.4123 0.1374 0.1740 0.1740 0.2340
openai_embed_3_small 0.3640 0.7281 0.2427 0.3026 0.3026 0.4097
ko-sroberta-multitask 0.2939 0.5877 0.1959 0.2500 0.2500 0.3351
openai_embed_3_large 0.3947 0.7895 0.2632 0.3348 0.3348 0.4491
KU-HIAI-ONTHEIT-large-v1 0.4386 0.8772 0.2924 0.3421 0.3421 0.4766
KU-HIAI-ONTHEIT-large-v1.1 0.4430 0.8860 0.2953 0.3406 0.3406 0.4778
kf-deberta-multitask 0.3158 0.6316 0.2105 0.2792 0.2792 0.3679
gte-multilingual-base 0.4035 0.8070 0.2690 0.3450 0.3450 0.4614
BGE-m3 0.4254 0.8508 0.2836 0.3421 0.3421 0.4701
BGE-m3-ko 0.4517 0.9035 0.3011 0.3494 0.3494 0.4886
  • Top-k 5
Model name F1 Recall Precision mAP mRR NDCG
paraphrase-multilingual-mpnet-base-v2 0.1813 0.5439 0.1088 0.1575 0.1575 0.2491
KoSimCSE-roberta 0.2164 0.6491 0.1298 0.1751 0.1751 0.2873
Cohere embed-multilingual-v3.0 0.2076 0.6228 0.1246 0.1640 0.1640 0.2731
openai ada 002 0.2602 0.7807 0.1561 0.2139 0.2139 0.3486
multilingual-e5-large-instruct 0.2544 0.7632 0.1526 0.2194 0.2194 0.3487
Upstage Embedding 0.2982 0.8947 0.1789 0.2237 0.2237 0.3822
paraphrase-multilingual-MiniLM-L12-v2 0.1637 0.4912 0.0982 0.1437 0.1437 0.2264
openai_embed_3_small 0.2690 0.8070 0.1614 0.2148 0.2148 0.3553
ko-sroberta-multitask 0.2164 0.6491 0.1298 0.1697 0.1697 0.2835
openai_embed_3_large 0.2807 0.8421 0.1684 0.2088 0.2088 0.3586
KU-HIAI-ONTHEIT-large-v1 0.3041 0.9123 0.1825 0.2137 0.2137 0.3783
KU-HIAI-ONTHEIT-large-v1.1 0.3099 0.9298 0.1860 0.2148 0.2148 0.3834
kf-deberta-multitask 0.2281 0.6842 0.1368 0.1724 0.1724 0.2939
gte-multilingual-base 0.2865 0.8596 0.1719 0.2096 0.2096 0.3637
BGE-m3 0.4254 0.8508 0.2836 0.3421 0.3421 0.4701
BGE-m3-ko 0.3099 0.9298 0.1860 0.2098 0.2098 0.3793
  • Top-k 10
Model name F1 Recall Precision mAP mRR NDCG
paraphrase-multilingual-mpnet-base-v2 0.1212 0.6667 0.0667 0.1197 0.1197 0.2382
KoSimCSE-roberta 0.1324 0.7281 0.0728 0.1080 0.1080 0.2411
Cohere embed-multilingual-v3.0 0.1324 0.7281 0.0728 0.1150 0.1150 0.2473
openai ada 002 0.1563 0.8596 0.0860 0.1051 0.1051 0.2673
multilingual-e5-large-instruct 0.1483 0.8158 0.0816 0.0980 0.0980 0.2520
Upstage Embedding 0.1707 0.9386 0.0939 0.1078 0.1078 0.2848
paraphrase-multilingual-MiniLM-L12-v2 0.1053 0.5789 0.0579 0.0961 0.0961 0.2006
openai_embed_3_small 0.1547 0.8509 0.0851 0.0984 0.0984 0.2593
ko-sroberta-multitask 0.1276 0.7018 0.0702 0.0986 0.0986 0.2275
openai_embed_3_large 0.1643 0.9035 0.0904 0.1180 0.1180 0.2855
KU-HIAI-ONTHEIT-large-v1 0.1707 0.9386 0.0939 0.1105 0.1105 0.2860
KU-HIAI-ONTHEIT-large-v1.1 0.1722 0.9474 0.0947 0.1033 0.1033 0.2822
kf-deberta-multitask 0.1388 0.7632 0.0763 0.1 0.1 0.2422
gte-multilingual-base 0.1675 0.9211 0.0921 0.1066 0.1066 0.2805
BGE-m3 0.4254 0.8508 0.2836 0.3421 0.3421 0.4701
BGE-m3-ko 0.1770 0.9736 0.0974 0.1097 0.1097 0.2932
Information Retrieval
Metric Value
cosine_accuracy@1 0.6103
cosine_accuracy@3 0.8169
cosine_accuracy@5 0.8732
cosine_accuracy@10 0.9202
cosine_precision@1 0.6103
cosine_precision@3 0.3787
cosine_precision@5 0.2761
cosine_precision@10 0.1728
cosine_recall@1 0.3847
cosine_recall@3 0.5902
cosine_recall@5 0.6794
cosine_recall@10 0.7695
cosine_ndcg@10 0.6723
cosine_mrr@10 0.7262
cosine_map@100 0.6074
dot_accuracy@1 0.6103
dot_accuracy@3 0.8169
dot_accuracy@5 0.8732
dot_accuracy@10 0.9202
dot_precision@1 0.6103
dot_precision@3 0.3787
dot_precision@5 0.2761
dot_precision@10 0.1728
dot_recall@1 0.3847
dot_recall@3 0.5902
dot_recall@5 0.6794
dot_recall@10 0.7695
dot_ndcg@10 0.6723
dot_mrr@10 0.7262
dot_map@100 0.6074
Bias, Risks and Limitations
  • Since the evaluation results are different for each domain, it is necessary to compare and evaluate the model in your own domain. In the Miracl benchmark, the evaluation was conducted using the Korean Wikipedia as a corpus, and in this case, the cosine_ndcg@10 score dropped by 0.2 points after learning. However, in the Auto-RAG benchmark, which is a financial domain, the ndcg score increased by 0.9 when it was top 1. This model may be advantageous for use in a specific domain.
  • Also, since the miracl benchmark consists of a corpus of relatively short strings, while the Korean Embedding Benchmark consists of a corpus of longer strings, this model may be more advantageous if the length of the corpus you want to use is long.
Training Hyperparameters
Non-Default Hyperparameters

The batch size was referenced from the following paper: Text Embeddings by Weakly-Supervised Contrastive Pre-training ( https://arxiv.org/pdf/2212.03533 )

  • eval_strategy : steps
  • per_device_train_batch_size : 32768
  • per_device_eval_batch_size : 32768
  • learning_rate : 3e-05
  • warmup_ratio : 0.03333333333333333
  • fp16 : True
  • batch_sampler : no_duplicates
All Hyperparameters
Click to expand
  • overwrite_output_dir : False
  • do_predict : False
  • eval_strategy : steps
  • prediction_loss_only : True
  • per_device_train_batch_size : 32768
  • per_device_eval_batch_size : 32768
  • per_gpu_train_batch_size : None
  • per_gpu_eval_batch_size : None
  • gradient_accumulation_steps : 1
  • eval_accumulation_steps : None
  • learning_rate : 3e-05
  • weight_decay : 0.0
  • adam_beta1 : 0.9
  • adam_beta2 : 0.999
  • adam_epsilon : 1e-08
  • max_grad_norm : 1.0
  • num_train_epochs : 3
  • max_steps : -1
  • lr_scheduler_type : linear
  • lr_scheduler_kwargs : {}
  • warmup_ratio : 0.03333333333333333
  • warmup_steps : 0
  • log_level : passive
  • log_level_replica : warning
  • log_on_each_node : True
  • logging_nan_inf_filter : True
  • save_safetensors : True
  • save_on_each_node : False
  • save_only_model : False
  • restore_callback_states_from_checkpoint : False
  • no_cuda : False
  • use_cpu : False
  • use_mps_device : False
  • seed : 42
  • data_seed : None
  • jit_mode_eval : False
  • use_ipex : False
  • bf16 : False
  • fp16 : True
  • fp16_opt_level : O1
  • half_precision_backend : auto
  • bf16_full_eval : False
  • fp16_full_eval : False
  • tf32 : None
  • local_rank : 0
  • ddp_backend : None
  • tpu_num_cores : None
  • tpu_metrics_debug : False
  • debug : []
  • dataloader_drop_last : True
  • dataloader_num_workers : 0
  • dataloader_prefetch_factor : None
  • past_index : -1
  • disable_tqdm : False
  • remove_unused_columns : True
  • label_names : None
  • load_best_model_at_end : False
  • ignore_data_skip : False
  • fsdp : []
  • fsdp_min_num_params : 0
  • fsdp_config : {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
  • fsdp_transformer_layer_cls_to_wrap : None
  • accelerator_config : {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
  • deepspeed : None
  • label_smoothing_factor : 0.0
  • optim : adamw_torch
  • optim_args : None
  • adafactor : False
  • group_by_length : False
  • length_column_name : length
  • ddp_find_unused_parameters : None
  • ddp_bucket_cap_mb : None
  • ddp_broadcast_buffers : False
  • dataloader_pin_memory : True
  • dataloader_persistent_workers : False
  • skip_memory_metrics : True
  • use_legacy_prediction_loop : False
  • push_to_hub : False
  • resume_from_checkpoint : None
  • hub_model_id : None
  • hub_strategy : every_save
  • hub_private_repo : False
  • hub_always_push : False
  • gradient_checkpointing : False
  • gradient_checkpointing_kwargs : None
  • include_inputs_for_metrics : False
  • eval_do_concat_batches : True
  • fp16_backend : auto
  • push_to_hub_model_id : None
  • push_to_hub_organization : None
  • mp_parameters :
  • auto_find_batch_size : False
  • full_determinism : False
  • torchdynamo : None
  • ray_scope : last
  • ddp_timeout : 1800
  • torch_compile : False
  • torch_compile_backend : None
  • torch_compile_mode : None
  • dispatch_batches : None
  • split_batches : None
  • include_tokens_per_second : False
  • include_num_input_tokens_seen : False
  • neftune_noise_alpha : None
  • optim_target_modules : None
  • batch_eval_metrics : False
  • batch_sampler : no_duplicates
  • multi_dataset_batch_sampler : proportional
Citation
BibTeX
Sentence Transformers
@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}
@misc{bge-m3,
      title={BGE M3-Embedding: Multi-Lingual, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation}, 
      author={Jianlv Chen and Shitao Xiao and Peitian Zhang and Kun Luo and Defu Lian and Zheng Liu},
      year={2024},
      eprint={2402.03216},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
@article{wang2022text,
  title={Text Embeddings by Weakly-Supervised Contrastive Pre-training},
  author={Wang, Liang and Yang, Nan and Huang, Xiaolong and Jiao, Binxing and Yang, Linjun and Jiang, Daxin and Majumder, Rangan and Wei, Furu},
  journal={arXiv preprint arXiv:2212.03533},
  year={2022}
}

Runs of dragonkue BGE-m3-ko on huggingface.co

76.5K
Total runs
-3.1K
24-hour runs
-4.8K
3-day runs
-6.7K
7-day runs
-183.8K
30-day runs

More Information About BGE-m3-ko huggingface.co Model

More BGE-m3-ko license Visit here:

https://choosealicense.com/licenses/apache-2.0

BGE-m3-ko huggingface.co

BGE-m3-ko huggingface.co is an AI model on huggingface.co that provides BGE-m3-ko's model effect (), which can be used instantly with this dragonkue BGE-m3-ko model. huggingface.co supports a free trial of the BGE-m3-ko model, and also provides paid use of the BGE-m3-ko. Support call BGE-m3-ko model through api, including Node.js, Python, http.

dragonkue BGE-m3-ko online free

BGE-m3-ko huggingface.co is an online trial and call api platform, which integrates BGE-m3-ko's modeling effects, including api services, and provides a free online trial of BGE-m3-ko, you can try BGE-m3-ko online for free by clicking the link below.

dragonkue BGE-m3-ko online free url in huggingface.co:

https://huggingface.co/dragonkue/BGE-m3-ko

BGE-m3-ko install

BGE-m3-ko is an open source model from GitHub that offers a free installation service, and any user can find BGE-m3-ko on GitHub to install. At the same time, huggingface.co provides the effect of BGE-m3-ko install, users can directly use BGE-m3-ko installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

BGE-m3-ko install url in huggingface.co:

https://huggingface.co/dragonkue/BGE-m3-ko

Url of BGE-m3-ko

Provider of BGE-m3-ko huggingface.co

dragonkue
ORGANIZATIONS

Other API from dragonkue