AlephBench
Models answer Hebrew reasoning, knowledge, and arithmetic questions. They also extract answers from passages and translate English into Hebrew.
The authors publish 11 task scores and an overall score for six models. The June release contains a May results snapshot.
- Made by
- HebArabNlpProject
- Released
- June 2026
- Kind
- Reasoning in many languages
- Level
- School
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Languages and literature, Mathematics
- Interaction
- Single turn
- Items from
- Another benchmark, Translated
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Gemma 4 31B IT #101 on the capability index
- Last model added
- April 2026
- Scores refreshed
- November 20th, 2026
Gemini 2.5 Flash leads with 88.8 among models released by April 2026 Gemini 2.5 Flash leads with 88.8 among models released by April 2026
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. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1
Gemini 2.5 Flash Sep 2025 88.8
- 2
Gemma 4 31B IT 88.1
- 3 dicta-il/DictaLM-3.0-24B-Thinking 85.2
- 4 gpt-oss-120b 84.2
- 5
Gemma 3 27B 83.5
- 6 HebArabNlpProject/Hebatron 77.1
Source: Official AlephBench leaderboard
- 1
Gemini 2.5 Flash Sep 2025 88.8
- 2
Gemma 4 31B IT 88.1
- 3 gpt-oss-120b 84.2
- 4
Gemma 3 27B 83.5
Source: Official AlephBench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 88.8 | Sep 2025 | Official leaderboard | |
| 2 | | 88.1 | Apr 2026 | Official leaderboard | |
| 3 | dicta-il/DictaLM-3.0-24B-Thinking | 85.2 | Official leaderboard | ||
| 4 | gpt-oss-120b | 84.2 | Aug 2025 | Official leaderboard | |
| 5 | | 83.5 | Mar 2025 | Official leaderboard | |
| 6 | HebArabNlpProject/Hebatron | 77.1 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 88.8 | Sep 2025 | Official leaderboard | |
| 2 | | 88.1 | Apr 2026 | Official leaderboard | |
| 3 | gpt-oss-120b | 84.2 | Aug 2025 | Official leaderboard | |
| 4 | | 83.5 | Mar 2025 | Official leaderboard |
Rankings by part
We read 11 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as MMLU, ARC and HellaSwag. Gemini 2.5 Flash leads 7 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
MMLU
4 models we track have a score here.
- 1 gpt-oss-120b ↑2 85.1
- 2
Gemini 2.5 Flash Sep 2025 ↓1 82.8
- 3
Gemma 4 31B IT ↓1 81.9
- 4
Gemma 3 27B 72.4
ARC
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 96.3
- 2
Gemma 4 31B IT 95.8
- 3 gpt-oss-120b 93.7
- 4
Gemma 3 27B 89.2
HellaSwag
4 models we track have a score here.
- 1
Gemma 4 31B IT ↑1 74.0
- 2
Gemini 2.5 Flash Sep 2025 ↓1 72.4
- 3
Gemma 3 27B ↑1 69.0
- 4 gpt-oss-120b ↓1 66.1
GSM8K
4 models we track have a score here.
- 1
Gemma 4 31B IT ↑1 95.9
- 2
Gemini 2.5 Flash Sep 2025 ↓1 95.2
- 3
Gemma 3 27B ↑1 92.3
- 4 gpt-oss-120b ↓1 91.8
COPA
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 95.2
- 2
Gemma 4 31B IT 94.2
- 3 gpt-oss-120b 91.0
- 4
Gemma 3 27B 90.8
Hebrew-QA
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 98.9
- 2
Gemma 4 31B IT 98.9
- 3
Gemma 3 27B ↑1 98.6
- 4 gpt-oss-120b ↓1 97.2
HebNLI
4 models we track have a score here.
- 1
Gemma 4 31B IT ↑1 85.2
- 2
Gemini 2.5 Flash Sep 2025 ↓1 82.5
- 3 gpt-oss-120b 79.5
- 4
Gemma 3 27B 73.6
Winograd
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 87.1
- 2
Gemma 4 31B IT 86.0
- 3
Gemma 3 27B ↑1 79.1
- 4 gpt-oss-120b ↓1 67.6
Sentiment
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 83.4
- 2 gpt-oss-120b ↑1 81.4
- 3
Gemma 4 31B IT ↓1 81.4
- 4
Gemma 3 27B 78.5
Trivia
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 96.0
- 2
Gemma 4 31B IT 89.4
- 3
Gemma 3 27B ↑1 88.9
- 4 gpt-oss-120b ↓1 88.4
Translation
4 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 87.0
- 2
Gemma 4 31B IT 86.1
- 3
Gemma 3 27B ↑1 85.7
- 4 gpt-oss-120b ↓1 84.8
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official AlephBench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official AlephBench 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
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 7 months ago, in April 2026, for Gemma 4 31B IT. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (6 entries).
- Pass Models scored 4 models we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 7 months ago, in April 2026, for Gemma 4 31B IT.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemma 4 31B IT (#101).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on huggingface.co: 4 rows for models we track and 2 other entries at the last read.
- 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.