Link

Model History

Table of contents


This page contains links to downloadable models for all historical versions of Stanza. Note that model versions prior to 0.2.0 (inclusive) require the StanfordNLP package (pip install stanfordnlp), and those after require the Stanza package (pip install stanza).

1.4.0

Upgraded English constituency parser and other conparse models. NER models for JA, MY

1.3.0

English constituency parser added. No charlm or bert integration yet, so the scores were quite low relative to state of the art. Multilingual langid model added.

1.2.3

NER models for AF, IT.

1.2.1

NER models for BG, HU, FI, VI, and all tokenize/mwt/lemma/pos/depparse models updated to UD 2.8.

1.1.1

This version extends Stanza’s v1.0.0 models with additional sentiment analysis models for English, German and Chinese pipelines, a new tokenizer for the Thai language, and new suite of Biomedical and Clinical English syntactic analysis and NER model packages. We’ve changed the default packages for a few languages for robustness. These changes include: Polish (default is now PDB model, from previous LFG), Korean (default is now GSD, from previous Kaist), Lithuanian (default is now ALKSNIS, from previous HSE).

1.0.0

For models used by Stanza v1.0.0 (latest version), you can find all related information on the Usage: Available Models page.

0.2.0

LanguageTreebankLanguage codeTreebank codeModels
AfrikaansAfriBoomsafaf_afribooms0.2.0
Ancient GreekPerseusgrcgrc_perseus0.2.0
 PROIELgrcgrc_proiel0.2.0
ArabicPADTarar_padt0.2.0
ArmenianArmTDPhyhy_armtdp0.2.0
BasqueBDTeueu_bdt0.2.0
BulgarianBTBbgbg_btb0.2.0
BuryatBDTbxrbxr_bdt0.2.0
CatalanAnCoracaca_ancora0.2.0
Chinese (traditional)GSDzhzh_gsd0.2.0
CroatianSEThrhr_set0.2.0
CzechCACcscs_cac0.2.0
 FicTreecscs_fictree0.2.0
 PDTcscs_pdt0.2.0
DanishDDTdada_ddt0.2.0
DutchAlpinonlnl_alpino0.2.0
 LassySmallnlnl_lassysmall0.2.0
EnglishEWTenen_ewt0.2.0
 GUMenen_gum0.2.0
 LinESenen_lines0.2.0
EstonianEDTetet_edt0.2.0
FinnishFTBfifi_ftb0.2.0
 TDTfifi_tdt0.2.0
FrenchGSDfrfr_gsd0.2.0
 Sequoiafrfr_sequoia0.2.0
 Spokenfrfr_spoken0.2.0
GalicianCTGglgl_ctg0.2.0
 TreeGalglgl_treegal0.2.0
GermanGSDdede_gsd0.2.0
GothicPROIELgotgot_proiel0.2.0
GreekGDTelel_gdt0.2.0
HebrewHTBhehe_htb0.2.0
HindiHDTBhihi_hdtb0.2.0
HungarianSzegedhuhu_szeged0.2.0
IndonesianGSDidid_gsd0.2.0
IrishIDTgaga_idt0.2.0
ItalianISDTitit_isdt0.2.0
 PoSTWITAitit_postwita0.2.0
JapaneseGSDjaja_gsd0.2.0
KazakhKTBkkkk_ktb0.2.0
KoreanGSDkoko_gsd0.2.0
 Kaistkoko_kaist0.2.0
KurmanjiMGkmrkmr_mg0.2.0
LatinITTBlala_ittb0.2.0
 Perseuslala_perseus0.2.0
 PROIELlala_proiel0.2.0
LatvianLVTBlvlv_lvtb0.2.0
North SamiGiellasmesme_giella0.2.0
NorwegianBokmaalno_bokmaalno_bokmaal0.2.0
 Nynorskno_nynorskno_nynorsk0.2.0
 NynorskLIAno_nynorskno_nynorsklia0.2.0
Old Church SlavonicPROIELcucu_proiel0.2.0
Old FrenchSRCMFfrofro_srcmf0.2.0
PersianSerajifafa_seraji0.2.0
PolishLFGplpl_lfg0.2.0
 SZplpl_sz0.2.0
PortugueseBosqueptpt_bosque0.2.0
RomanianRRTroro_rrt0.2.0
RussianSynTagRusruru_syntagrus0.2.0
 Taigaruru_taiga0.2.0
SerbianSETsrsr_set0.2.0
SlovakSNKsksk_snk0.2.0
SlovenianSSJslsl_ssj0.2.0
 SSTslsl_sst0.2.0
SpanishAnCoraeses_ancora0.2.0
SwedishLinESsvsv_lines0.2.0
 Talbankensvsv_talbanken0.2.0
TurkishIMSTtrtr_imst0.2.0
UkrainianIUukuk_iu0.2.0
Upper SorbianUFALhsbhsb_ufal0.2.0
UrduUDTBurur_udtb0.2.0
UyghurUDTugug_udt0.2.0
VietnameseVTBvivi_vtb0.2.0

0.1.0

LanguageTreebankLanguage codeTreebank codeModels
AfrikaansAfriBoomsafaf_afribooms0.1.0
Ancient GreekPerseusgrcgrc_perseus0.1.0
 PROIELgrcgrc_proiel0.1.0
ArabicPADTarar_padt0.1.0
ArmenianArmTDPhyhy_armtdp0.1.0
BasqueBDTeueu_bdt0.1.0
BulgarianBTBbgbg_btb0.1.0
BuryatBDTbxrbxr_bdt0.1.0
CatalanAnCoracaca_ancora0.1.0
Chinese (traditional)GSDzhzh_gsd0.1.0
CroatianSEThrhr_set0.1.0
CzechCACcscs_cac0.1.0
 FicTreecscs_fictree0.1.0
 PDTcscs_pdt0.1.0
DanishDDTdada_ddt0.1.0
DutchAlpinonlnl_alpino0.1.0
 LassySmallnlnl_lassysmall0.1.0
EnglishEWTenen_ewt0.1.0
 GUMenen_gum0.1.0
 LinESenen_lines0.1.0
EstonianEDTetet_edt0.1.0
FinnishFTBfifi_ftb0.1.0
 TDTfifi_tdt0.1.0
FrenchGSDfrfr_gsd0.1.0
 Sequoiafrfr_sequoia0.1.0
 Spokenfrfr_spoken0.1.0
GalicianCTGglgl_ctg0.1.0
 TreeGalglgl_treegal0.1.0
GermanGSDdede_gsd0.1.0
GothicPROIELgotgot_proiel0.1.0
GreekGDTelel_gdt0.1.0
HebrewHTBhehe_htb0.1.0
HindiHDTBhihi_hdtb0.1.0
HungarianSzegedhuhu_szeged0.1.0
IndonesianGSDidid_gsd0.1.0
IrishIDTgaga_idt0.1.0
ItalianISDTitit_isdt0.1.0
 PoSTWITAitit_postwita0.1.0
JapaneseGSDjaja_gsd0.1.0
KazakhKTBkkkk_ktb0.1.0
KoreanGSDkoko_gsd0.1.0
 Kaistkoko_kaist0.1.0
KurmanjiMGkmrkmr_mg0.1.0
LatinITTBlala_ittb0.1.0
 Perseuslala_perseus0.1.0
 PROIELlala_proiel0.1.0
LatvianLVTBlvlv_lvtb0.1.0
North SamiGiellasmesme_giella0.1.0
NorwegianBokmaalno_bokmaalno_bokmaal0.1.0
 Nynorskno_nynorskno_nynorsk0.1.0
 NynorskLIAno_nynorskno_nynorsklia0.1.0
Old Church SlavonicPROIELcucu_proiel0.1.0
Old FrenchSRCMFfrofro_srcmf0.1.0
PersianSerajifafa_seraji0.1.0
PolishLFGplpl_lfg0.1.0
 SZplpl_sz0.1.0
PortugueseBosqueptpt_bosque0.1.0
RomanianRRTroro_rrt0.1.0
RussianSynTagRusruru_syntagrus0.1.0
 Taigaruru_taiga0.1.0
SerbianSETsrsr_set0.1.0
SlovakSNKsksk_snk0.1.0
SlovenianSSJslsl_ssj0.1.0
 SSTslsl_sst0.1.0
SpanishAnCoraeses_ancora0.1.0
SwedishLinESsvsv_lines0.1.0
 Talbankensvsv_talbanken0.1.0
TurkishIMSTtrtr_imst0.1.0
UkrainianIUukuk_iu0.1.0
Upper SorbianUFALhsbhsb_ufal0.1.0
UrduUDTBurur_udtb0.1.0
UyghurUDTugug_udt0.1.0
VietnameseVTBvivi_vtb0.1.0