ParsiNLU
Models read Persian passages, compare queries, and identify sentiment. The six tasks also cover entailment, multiple-choice questions, and translation.
The project has 171 GitHub stars. Mofid AI reports separate model scores for its literary translation test in both directions.
- Made by
- Daniel Khashabi and colleagues
- Released
- October 2020
- Kind
- Reading in many languages
- Level
- University
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Languages and literature
- Interaction
- Single turn
- Items from
- Written by crowdworkers, Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
mT5 Per + Eng (XL) tops the leaderboard with 74.7
Best score per entry from the benchmark's paper, read on November 21st, 2026. A paper’s table doesn’t add newer models. The paper reports no overall score, so this is the multiple-choice all column, which it presents first. 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.
- 74.7 mT5 Per + Eng (XL)
- 73.6 mT5 Per + Eng (large)
- 70.4 mT5 (XL)
- 68.2 mT5 English (XL)
- 67.4 mT5 English (large)
- 63.9 mT5 Per + Eng (base)
- 53.4 mT5 English (base)
- 49.2 mT5 (large)
- 49.0 mBERT (base)
- 45.3 mT5 Per + Eng (small)
- 42.6 mT5 (base)
- 40.7 ParsBERT (base)
- 39.2 WikiBERT (base)
- 33.0 mT5 English (small)
- 30.9 mT5 (small)
- 1 mT5 Per + Eng (XL) 74.7
- 2 mT5 Per + Eng (large) 73.6
- 3 mT5 (XL) 70.4
- 4 mT5 English (XL) 68.2
- 5 mT5 English (large) 67.4
- 6 mT5 Per + Eng (base) 63.9
- 7 mT5 English (base) 53.4
- 8 mT5 (large) 49.2
- 9 mBERT (base) 49.0
- 10 mT5 Per + Eng (small) 45.3
- 11 mT5 (base) 42.6
- 12 ParsBERT (base) 40.7
- 13 WikiBERT (base) 39.2
- 14 mT5 English (small) 33.0
- 15 mT5 (small) 30.9
Source: The ParsiNLU paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | mT5 Per + Eng (XL) | 74.7 | Official leaderboard | ||
| 2 | mT5 Per + Eng (large) | 73.6 | Official leaderboard | ||
| 3 | mT5 (XL) | 70.4 | Official leaderboard | ||
| 4 | mT5 English (XL) | 68.2 | Official leaderboard | ||
| 5 | mT5 English (large) | 67.4 | Official leaderboard | ||
| 6 | mT5 Per + Eng (base) | 63.9 | Official leaderboard | ||
| 7 | mT5 English (base) | 53.4 | Official leaderboard | ||
| 8 | mT5 (large) | 49.2 | Official leaderboard | ||
| 9 | mBERT (base) | 49.0 | Official leaderboard | ||
| 10 | mT5 Per + Eng (small) | 45.3 | Official leaderboard | ||
| 11 | mT5 (base) | 42.6 | Official leaderboard | ||
| 12 | ParsBERT (base) | 40.7 | Official leaderboard | ||
| 13 | WikiBERT (base) | 39.2 | Official leaderboard | ||
| 14 | mT5 English (small) | 33.0 | Official leaderboard | ||
| 15 | mT5 (small) | 30.9 | 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 benchmark’s paper (15 entries).
- Pass Models scored 15 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 aclanthology.org: 15 entries at the last read, none a model we track.
- Pass Last read November 21st, 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.