SweWinogender 2.0
Models judge whether a Swedish pronoun refers to a person in a stated occupation. The parity score checks agreement across three pronoun variants.
The official leaderboard scores eight pretrained encoders on 624 examples. Markovic reports accuracy for 23 models on 120 examples.
- Made by
- Språkbanken Text, University of Gothenburg
- Released
- February 2023
- Kind
- Bias, fairness and politics
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Propensity
- Behavior
- Bias
- Language
- Swedish
- Interaction
- Single turn
- Items from
- Another benchmark
- Refresh
- New versions replace it
- Run by
- Anyone
- Based on
- Winogender
KBLab/megatron-bert-large-swedish-cased-165k tops the leaderboard with -0.29
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.
- 1 KBLab/megatron-bert-large-swedish-cased-165k -0.29
- 2 xlm-roberta-base -0.30
- 3 xlm-roberta-large -0.32
- 4 KBLab/megatron-bert-base-swedish-cased-600k -0.32
- 5 KBLab/bert-base-swedish-cased-new -0.33
- 6 AI-Nordics/bert-large-swedish-cased -0.33
- 7 KB/bert-base-swedish-cased -0.33
- 8 NbAiLab/nb-bert-base -0.33
Source: Official SweWinogender 2.0 leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | KBLab/megatron-bert-large-swedish-cased-165k 0.4B | -0.29 | Official leaderboard | ||
| 2 | xlm-roberta-base 0.3B | -0.30 | Official leaderboard | ||
| 3 | xlm-roberta-large 0.6B | -0.32 | Official leaderboard | ||
| 4 | KBLab/megatron-bert-base-swedish-cased-600k 0.1B | -0.32 | Official leaderboard | ||
| 5 | KBLab/bert-base-swedish-cased-new 0.1B | -0.33 | Official leaderboard | ||
| 6 | AI-Nordics/bert-large-swedish-cased 0.3B | -0.33 | Official leaderboard | ||
| 7 | KB/bert-base-swedish-cased 0.1B | -0.33 | Official leaderboard | ||
| 8 | NbAiLab/nb-bert-base 0.2B | -0.33 | 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 Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6%
- Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts.
Gemini 4 Carbon 0%
- Covert misbehavior (Thinking Machines) In use Agentic tasks that check whether a model misbehaves and conceals it. Scored by the rate of covert misbehavior.
Inkling 2 Mini 8.30%
- Destructive actions (Thinking Machines) In use Agentic tasks that check whether a model takes destructive actions it was not authorized to take. Scored by the rate of destructive actions.
Inkling 2 Small 4.70%
- Misbehavior (Thinking Machines) In use Agentic tasks that check whether a model fabricates results, ignores instructions or hides what it did when a task is hard. Scored by the rate of misbehavior.
Inkling 2 Mini 9.80%
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 (8 entries).
- Pass Models scored 8 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: 8 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.