RACE-H
Models answer multiple-choice reading questions from English exams for Chinese high school students. The test split has 3,498 questions, scored by accuracy.
In the authors’ words
We present RACE, a new dataset for benchmark evaluation of methods in the reading comprehension task. Collected from the English exams for middle and high school Chinese students in the age range between 12 to 18, RACE consists of near 28,000 passages and near 100,000 questions generated by human experts (English instructors), and covers a variety of topics which are carefully designed for evaluating the students' ability in understanding and reasoning. In particular, the proportion of questions that requires reasoning is much larger in RACE than that in other benchmark datasets for reading comprehension, and there is a significant gap between the performance of the state-of-the-art models (43%) and the ceiling human performance (95%). We hope this new dataset can serve as a valuable resource for research and evaluation in machine comprehension. The dataset is freely available at http://www.cs.cmu.edu/~glai1/data/race/ and the code is available at https://github.com/qizhex/RACE_AR_baselines.
- Level
- School Students, from grade school to high school
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Multiple choice
Claude 3 Opus leads with 92.9% among models released by Dec 2024
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
The newest model with a score here came out in Dec 2024, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6 Astra
- Claude Opus 5.5
- GPT-6.1 Terra
- GPT-6.1 Sol
- and 16 more
- 1 Claude 3 Opus 92.9%
- 2 Claude 3 Sonnet 88.8%
- 3 Claude 2 88.3%
- 4 Claude 3 Haiku 87%
- 5 PaLM 2-L 62.3%
- 6
DeepSeek LLM 67B 56% - 7 DeepSeek-V3 51.3%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3 Opus | 92.9% | Mar 2024 | Model card | |
| 2 | Claude 3 Sonnet | 88.8% | Mar 2024 | Model card | |
| 3 | Claude 2 | 88.3% | Jul 2023 | Model card | |
| 4 | Claude 3 Haiku | 87% | Mar 2024 | Model card | |
| 5 | PaLM 2-L | 62.3% | May 2023 | Model card | |
| 6 | DeepSeek LLM 67B | 56% | Jan 2024 | Model card | |
| 7 | DeepSeek-V3 | 51.3% | Dec 2024 | Model card |
3 models have held number one since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
The best score passed 85% in Jul 2023
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs dropped it from launch posts once the best score passed 85%
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| 26 Dec 2024 | DeepSeek-V3 | 51.3 | RACE-High (Acc.) |
| 4 Mar 2024 | Claude 3 Opus | 92.9 | RACE-H (5-shot) |
| 11 Jul 2023 | Claude 2 | 88.3 | RACE-H (5-shot) |
| 10 May 2023 |  PaLM 2-L | 62.3 | RACE-H, 1-shot |
5 system cards report it, from 2 labs
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 4 Mar 2024 | Claude 3 Sonnet model card | 88.8 | RACE-H, 5-shot (Table 1) | Capabilities |
| 4 Mar 2024 | Claude 3 Opus model card | 92.9 | RACE-H, 5-shot (Table 1) | Capabilities |
| 4 Mar 2024 | Claude 3 Haiku model card | 87 | RACE-H, 5-shot (Table 1) | Capabilities |
| 5 Jan 2024 | DeepSeek LLM 67B technical report | 56 | RACE-High, DeepSeek 67B Chat (Table 6) | Capabilities |
| 11 Jul 2023 | Claude 2 model card | 88.3 | RACE-H (5-shot) (4.3) | Capabilities |
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% - 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% - KMMLU-Pro In use Models answer multiple-choice questions taken from Korean National Professional Licensure exams. It tests professional knowledge used in Korean industry.
Solar Pro 4 79.2% - BIG-Bench Extra Hard In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is a harmonic mean of task accuracies. Gemma 4 31B IT 74.4%
- HLE-Verified In use Verified and revised version of Humanity's Last Exam (1,811 items) with uncertain items removed.
Seed2.0 Pro 73.6%