SweWinograd 2.0
Models decide whether a highlighted Swedish pronoun refers to a proposed person or thing in the passage.
Markovic reports accuracy for 23 models on all 140 test items. The test contains 43 positive cases and 97 negative cases.
- Made by
- Språkbanken Text, University of Gothenburg
- Released
- February 2023
- Kind
- Reading in many languages
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Language
- Swedish
- Interaction
- Single turn
- Items from
- Written by experts
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- Winograd
AI-Nordics/bert-large-swedish-cased tops the leaderboard with 0.19
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 0.19 AI-Nordics/bert-large-swedish-cased
- 0.19 KBLab/megatron-bert-large-swedish-cased-165k
- 0.14 KB/bert-base-swedish-cased
- 0.12 NbAiLab/nb-bert-base
- 0.08 Random
- 0.08 xlm-roberta-large
- 0.06 KBLab/megatron-bert-base-swedish-cased-600k
- 0.06 SVM
- 0.04 KBLab/bert-base-swedish-cased-new
- -0.18 MaxFreq/Avg
- -0.18 Random Forest
- -0.18 xlm-roberta-base
- -0.24 Decision Tree
- 1 AI-Nordics/bert-large-swedish-cased 0.19
- 2 KBLab/megatron-bert-large-swedish-cased-165k 0.19
- 3 KB/bert-base-swedish-cased 0.14
- 4 NbAiLab/nb-bert-base 0.12
- 5 Random 0.08
- 6 xlm-roberta-large 0.08
- 7 KBLab/megatron-bert-base-swedish-cased-600k 0.06
- 8 SVM 0.06
- 9 KBLab/bert-base-swedish-cased-new 0.04
- 10 MaxFreq/Avg -0.18
- 11 Random Forest -0.18
- 12 xlm-roberta-base -0.18
- 13 Decision Tree -0.24
Source: Official SweWinograd 2.0 leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | AI-Nordics/bert-large-swedish-cased 0.3B | 0.19 | Official leaderboard | ||
| 2 | KBLab/megatron-bert-large-swedish-cased-165k 0.4B | 0.19 | Official leaderboard | ||
| 3 | KB/bert-base-swedish-cased 0.1B | 0.14 | Official leaderboard | ||
| 4 | NbAiLab/nb-bert-base 0.2B | 0.12 | Official leaderboard | ||
| 5 | Random | 0.08 | Official leaderboard | ||
| 6 | xlm-roberta-large 0.6B | 0.08 | Official leaderboard | ||
| 7 | KBLab/megatron-bert-base-swedish-cased-600k 0.1B | 0.06 | Official leaderboard | ||
| 8 | SVM | 0.06 | Official leaderboard | ||
| 9 | KBLab/bert-base-swedish-cased-new 0.1B | 0.04 | Official leaderboard | ||
| 10 | MaxFreq/Avg | -0.18 | Official leaderboard | ||
| 11 | Random Forest | -0.18 | Official leaderboard | ||
| 12 | xlm-roberta-base 0.3B | -0.18 | Official leaderboard | ||
| 13 | Decision Tree | -0.24 | Official leaderboard |
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Multilingual
- MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
- WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET.
K-EXAONE 90.5%
- MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language.
Qwen3.7-Max 87%
- CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories.
Solar Open 2 90.7%
- Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
- MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%
Data audit
All 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (13 entries).
- Pass Models scored 13 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on lab.kb.se: 13 entries at the last read, none a model we track.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.