CultureAtlas
Models judge whether statements about a region or cultural group are true. The authors pair statements drawn from Wikipedia with false alternatives and report F1.
The CulDPD authors trained on other datasets, then tested their models on 6,238 CultureAtlas statements.
- Released
- February 2024
- Kind
- Commonsense
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Sociology
- Language
- English
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Llama 2-13B #223 on the capability index
- Last model added
- July 2023
- Scores refreshed
- November 20th, 2026
ChatGPT 20B tops the leaderboard with 93.1% Vicuna-13B-v1.3 leads with 73.7% among models released by July 2023
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. 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. The best model we track is Vicuna-13B-v1.3, at 73.7%. This view shows only models we track, each at its best.
The newest model with a score here came out in July 2023, 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 ChatGPT 20B chat-HF 93.1%
- 2
Vicuna-13B-v1.3 chat-HF 73.7%
- 3
Llama 2-13B chat 69.7%
- 4 Vicuna 7B chat-HF 66.3%
- 5
Llama 2-7B chat 56.1%
Source: The CultureAtlas paper
- 1
Vicuna-13B-v1.3 chat-HF 73.7%
- 2
Llama 2-13B chat 69.7%
- 3
Llama 2-7B chat 56.1%
Source: The CultureAtlas paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | ChatGPT 20B | 93.1% | chat-HF | Official leaderboard | |
| 2 | | 73.7% | chat-HF | Jun 2023 | Official leaderboard |
| 3 | | 69.7% | chat | Jul 2023 | Official leaderboard |
| 4 | Vicuna 7B | 66.3% | chat-HF | Official leaderboard | |
| 5 | | 56.1% | chat | Jul 2023 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 73.7% | chat-HF | Jun 2023 | Official leaderboard |
| 2 | | 69.7% | chat | Jul 2023 | Official leaderboard |
| 3 | | 56.1% | chat | Jul 2023 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s paper: High Resource, Mid Resource and Low Resource. Each ranks models on one part of the test. Vicuna-13B-v1.3 leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
High Resource
3 models we track have a score here.
- 1
Vicuna-13B-v1.3 74.5%
- 2
Llama 2-13B 66.3%
- 3
Llama 2-7B 59.8%
Mid Resource
3 models we track have a score here.
- 1
Vicuna-13B-v1.3 75.3%
- 2
Llama 2-13B 69.3%
- 3
Llama 2-7B 56.6%
Low Resource
3 models we track have a score here.
- 1
Vicuna-13B-v1.3 74.3%
- 2
Llama 2-7B ↑1 39.2%
- 3
Llama 2-13B ↓1 31.9%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The CultureAtlas paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official CultureAtlas 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 Knowledge
- MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects.
Intern-S2-397B 89.8%
- MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy.
Seed1.8 92.3%
- SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy.
Qwen 3.6 Max (Preview) 73.9%
- WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1.
Tencent Hy4 preview 83.9%
- BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly.
Gemma 4 31B IT 74.4%
- HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in July 2023, for Llama 2-13B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the benchmark’s paper (5 entries).
- Pass Models scored 3 models we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 3 years ago, in July 2023, for Llama 2-13B.
- Note Frontier models None of the top 10 has a score. The strongest tested is Llama 2-13B (#223).
- Pass Headroom The best is 73.7%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 5 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.