{"id":21529594,"url":"https://github.com/sooftware/korean-plm","last_synced_at":"2026-02-05T21:04:16.331Z","repository":{"id":45001243,"uuid":"513220095","full_name":"sooftware/Korean-PLM","owner":"sooftware","description":"List of Korean pre-trained language models.","archived":false,"fork":false,"pushed_at":"2023-08-31T15:02:37.000Z","size":27,"stargazers_count":186,"open_issues_count":0,"forks_count":16,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-07-24T11:49:53.214Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/sooftware.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2022-07-12T16:43:27.000Z","updated_at":"2025-06-19T09:28:31.000Z","dependencies_parsed_at":"2025-05-26T16:16:29.886Z","dependency_job_id":null,"html_url":"https://github.com/sooftware/Korean-PLM","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/sooftware/Korean-PLM","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sooftware%2FKorean-PLM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sooftware%2FKorean-PLM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sooftware%2FKorean-PLM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sooftware%2FKorean-PLM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/sooftware","download_url":"https://codeload.github.com/sooftware/Korean-PLM/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/sooftware%2FKorean-PLM/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29134249,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-05T20:50:26.975Z","status":"ssl_error","status_checked_at":"2026-02-05T20:49:26.082Z","response_time":65,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-24T01:58:17.411Z","updated_at":"2026-02-05T21:04:13.947Z","avatar_url":"https://github.com/sooftware.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Korean-PLM (Korean Pre-trained Language Models)\n\n공개된 한국어 사전학습 모델을 기록합니다. 크게 아래 3개의 모델 계열로 구분했으며, 모델 사이즈는 정확하지 않을 수 있습니다.\n\n- Encoder Model (BERT 계열)\n- Decoder Model (GPT 계열)\n- Encoder-Decoder Model (Seq2seq 계열)\n  \n  \n## Encoder Model (BERT 계열)\n\n| Model Name     | Size           |  Link           |\n| :------------: | :------------: |  :------------: |\n|   LASSL BERT   |   Small (10M)   |      [link](https://huggingface.co/lassl/bert-ko-small)   |\n|   LASSL RoBERTa   |   Small (10M)   |      [link](https://huggingface.co/lassl/roberta-ko-small)   |\n|   TUNiB ELECTRA (Ko)   |   Small (10M)   |      [link](https://huggingface.co/tunib/electra-ko-small)   |\n|   TUNiB ELECTRA (Ko-En)   |   Small (10M)   |      [link](https://huggingface.co/tunib/electra-ko-en-small)   |\n|   KoELECTRA   |   Small (10M)   |      [link](https://huggingface.co/monologg/koelectra-small-v3-discriminator)   |\n|  KcELECTRA   |   Small (10M)   |     [link](https://huggingface.co/beomi/kcelectra-v2022-dev)   |\n|   Soongsil-BERT   |   Small (10M)   |   [link](https://huggingface.co/jason9693/soongsil-bert-small)   |\n|   Ko-CHAR-ELECTRA  |   Small (10M)   |     [link](https://huggingface.co/monologg/kocharelectra-small-discriminator)   |\n|   Distill-KoBERT   |   30M   |     [link](https://huggingface.co/monologg/kocharelectra-base-discriminator)   |\n|   KoBERT   |   Base (110M)   |      [link](https://huggingface.co/skt/kobert-base-v1)   |\n|   KorBERT   |   Base (110M)   |     [link](https://aiopen.etri.re.kr/service_dataset.php)   |\n|   HanBERT   |   Base (110M)  |     [link](https://github.com/monologg/HanBert-Transformers)   |\n| KcBERT    |   Base (110M)   |     [link](https://huggingface.co/beomi/kcbert-base)   |\n|   Soongsil-BERT   |   Base (110M)   |   [link](https://huggingface.co/jason9693/soongsil-bert-base)   |\n|   KLUE-RoBERTa   |   Base (110M)  |     [link](https://huggingface.co/klue/roberta-base)   |\n|   KoELECTRA   |   Base (110M)   |     [link](https://huggingface.co/monologg/koelectra-base-v3-discriminator)   |\n|   KcELECTRA   |   Base (110M)   |     [link](https://huggingface.co/beomi/KcELECTRA-base)   |\n|   TUNiB-ELECTRA (Ko)   |   Base (110M)   |      [link](https://huggingface.co/tunib/electra-ko-base)   |\n|   TUNiB-ELECTRA (Ko-En)   |   Base (110M)   |     [link](https://huggingface.co/tunib/electra-ko-en-base)   |\n|   LMKor Albert   |   Base (110M)   |      [link](https://huggingface.co/kykim/albert-kor-base)   |\n|   LMKor Bert   |   Base (110M)  |    [link](https://huggingface.co/kykim/bert-kor-base)   |\n|   LMKor Funnel   |   Base (110M)   |    [link](https://huggingface.co/kykim/funnel-kor-base)   |\n|   LMKor ELECTRA   |   Base (110M)   |   [link](https://huggingface.co/kykim/electra-kor-base)   |\n|   KalBERT   |   Base (110M)   |     [link](https://github.com/MrBananaHuman/KalBert)   |\n|   LASSL BERT   |   Base (110M)   |      [link](https://huggingface.co/lassl/bert-ko-base)   |\n|   KoBigBird   |   Base (110M)   |      [link](https://huggingface.co/monologg/kobigbird-bert-base)   |\n|   Ko-CHAR-BERT   |   Base (110M)   |     [link](https://github.com/MrBananaHuman/KoreanCharacterBert)   |\n|   Ko-CHAR-ELECTRA   |   Base (110M)   |     [link](https://huggingface.co/monologg/kocharelectra-base-discriminator)   |\n|   BrainSBERT   |   Base (110M)   |     [link](https://huggingface.co/hyunwoongko/brainsbert-base)   |\n| KPFBERT | Base (110M) | [link](https://huggingface.co/jinmang2/kpfbert) |\n| KcBERT     |   Large (340M)   |     [link](https://huggingface.co/beomi/kcbert-large)   |\n|   KLUE-RoBERTa   |   Large (340M)  |   [link](https://huggingface.co/klue/roberta-large)   |\n\n## Decoder Model (GPT 계열)\n\n| Model Name     | Size           |  Link           |\n| :------------: | :------------: |  :------------: |\n|   LASSL GPT2   |   Small (20M)   |      [link](https://huggingface.co/lassl/gpt2-ko-small)   |\n|   LMKor KoGPT2   |   Base (110M)   |      [link](https://huggingface.co/kykim/gpt3-kor-small_based_on_gpt2)   |\n|   SKT KoGPT2   |   Base (110M)   |      [link](https://huggingface.co/skt/kogpt2-base-v2)   |\n|   SKT KoGPT Trinity   |   1.2B   |      [link](https://huggingface.co/skt/ko-gpt-trinity-1.2B-v0.5)   |\n|   KakaoBrain KoGPT   |   6B   |      [link](https://huggingface.co/kakaobrain/kogpt)   |\n|   EleutherAI Polyglot-Ko   |   1.3B   |      [link](https://huggingface.co/EleutherAI/polyglot-ko-1.3b)   |\n|   EleutherAI Polyglot-Ko   |   3.8B   |      [link](EleutherAI/polyglot-ko-3.8b)   |\n|   EleutherAI Polyglot-Ko   |   5.8B   |      [link](https://huggingface.co/EleutherAI/polyglot-ko-5.8b)   |\n|   EleutherAI Polyglot-Ko   |   12.8B   |      [link](https://huggingface.co/EleutherAI/polyglot-ko-12.8b)   |\n|   KoAlpaca (Polyglot-Ko)   |   5.8B   |      [link](https://huggingface.co/beomi/KoAlpaca-Polyglot-5.8B)   |\n|   KoAlpaca (Polyglot-Ko)   |   12.8B   |      [link](https://huggingface.co/beomi/KoAlpaca-Polyglot-12.8B)   |\n\n## Encoder-Decoder Model (Seq2seq 계열)  \n\n| Model Name     | Size           |  Link           |\n| :------------: | :------------: |  :------------: |\n|   Cosmoquester BART   |   Mini (13M)   |      [link](https://huggingface.co/cosmoquester/bart-ko-mini)   |\n|   KE-T5 (Ko-En)   |   Small (60M)   |      [link](https://huggingface.co/KETI-AIR/ke-t5-small)   |\n|   KE-T5 (Ko)   |   Small (60M)   |      [link](https://huggingface.co/KETI-AIR/ke-t5-small-ko)   |\n|   LMKor T5  |   Small (60M)   |     [link](https://huggingface.co/kykim/t5-kor-small)   |\n|   Cosmoquester BART   |   Small (40M)   |      [link](https://huggingface.co/cosmoquester/bart-ko-small)   |\n|   LMKor BERT Shared   |   Base (130M)   |      [link](https://huggingface.co/kykim/bertshared-kor-base)   |\n|   SKT KoBART   |   Base (130M)   |      [link](https://huggingface.co/gogamza/kobart-base-v2)   |\n|   Cosmoquester BART   |   Base (130M)   |      [link](https://huggingface.co/gogamza/kobart-base-v2)   |\n| Kolang-T5     |   Base (220M)   |     [link](https://huggingface.co/digit82/kolang-t5-base)   |\n| KcT5     |   Base (220M)   |     [link](https://huggingface.co/beomi/KcT5-dev)   |\n| KE-T5 (Ko-En)    |   Base (220M)   |     [link](https://huggingface.co/KETI-AIR/ke-t5-base)   |\n| KE-T5 (Ko)    |   Base (220M)   |     [link](https://huggingface.co/KETI-AIR/ke-t5-basko)   |\n|   Asian BART (Ko)   |   Base (360M)   |      [link](hyunwoongko/asian-bart-ko)   |\n| pko-t5 (Ko) | Base (250M) | [link](https://huggingface.co/paust/pko-t5-base) |\n| Wisenut KoT5 (Ko) | Base (220M) | [link](https://github.com/wisenut-research/KoT5#usage) |\n|   mBART   |   Large (610M)   |      [link](https://huggingface.co/facebook/mbart-large-50)   |\n| KE-T5 (Ko-En)    |   Large (770M)   |     [link](https://huggingface.co/KETI-AIR/ke-t5-large)   |\n| KE-T5 (Ko)    |   Large (770M)   |     [link](https://huggingface.co/KETI-AIR/ke-t5-large-ko)   |\n| pko-t5 (Ko) | Large (800M) | [link](https://huggingface.co/paust/pko-t5-large) |\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsooftware%2Fkorean-plm","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsooftware%2Fkorean-plm","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsooftware%2Fkorean-plm/lists"}