CLUEWSC
DeepSeek-R1 92.8% leads since Jan 2025 Fading Models decide whether a pronoun in a Chinese sentence refers to a given noun phrase, a Winograd-style task. Scored by accuracy.
Duplicate of cluewsc: same CLUE Winograd task and same paper. Merge into cluewsc and drop this record.
In the authors’ words
The advent of natural language understanding (NLU) benchmarks for English, such as GLUE and SuperGLUE allows new NLU models to be evaluated across a diverse set of tasks. These comprehensive benchmarks have facilitated a broad range of research and applications in natural language processing (NLP). The problem, however, is that most such benchmarks are limited to English, which has made it difficult to replicate many of the successes in English NLU for other languages. To help remedy this issue, we introduce the first large-scale Chinese Language Understanding Evaluation (CLUE) benchmark. CLUE is an open-ended, community-driven project that brings together 9 tasks spanning several well-established single-sentence/sentence-pair classification tasks, as well as machine reading comprehension, all on original Chinese text. To establish results on these tasks, we report scores using an exhaustive set of current state-of-the-art pre-trained Chinese models (9 in total). We also introduce a number of supplementary datasets and additional tools to help facilitate further progress on Chinese NLU. Our benchmark is released at https://www.CLUEbenchmarks.com
- Level
- Everyday Most adults, without training
- Scored
- Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
- Answers
- Multiple choice
DeepSeek-R1 leads with 92.8% among models released by Apr 2026
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 Apr 2026, 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 DeepSeek-R1 92.8%
- 2 Kimi k1.5 91.7%
- 3 DeepSeek-V3 90.9%
- 4 DeepSeek-V2 (MoE-236B, May 2024) 89.9%
- 5 DeepSeek-V4-Pro 85.2%
- 6 GLM-5 84.2%
- 7 GLM-4.5 83.5%
- 8 DeepSeek LLM 67B 60%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | DeepSeek-R1 | 92.8% | Jan 2025 | Model card | |
| 2 | Kimi k1.5 | 91.7% | Jan 2025 | Model card | |
| 3 | DeepSeek-V3 | 90.9% | Dec 2024 | Model card | |
| 4 | DeepSeek-V2 (MoE-236B, May 2024) | 89.9% | May 2024 | Model card | |
| 5 | DeepSeek-V4-Pro | 85.2% | Apr 2026 | Model card | |
| 6 | GLM-5 | 84.2% | Feb 2026 | Model card | |
| 7 | GLM-4.5 | 83.5% | Aug 2025 | Model card | |
| 8 | DeepSeek LLM 67B | 60% | Jan 2024 | Model card |
4 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 May 2024
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Labs headline it less often now, last in Apr 2026
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 |
|---|---|---|---|
| 24 Apr 2026 | DeepSeek-V4-Pro | 85.2 | CLUEWSC (EM) |
| 16 Mar 2025 |  ERNIE 4.5 | CLUEWSC | |
| 20 Jan 2025 | Kimi k1.5 | 91.7 | CLUEWSC (EM), short-CoT |
| 20 Jan 2025 | DeepSeek-R1 | 92.8 | CLUEWSC (EM) |
| 26 Dec 2024 | DeepSeek-V3 | 90.9 | CLUEWSC (EM) |
7 system cards report it, from 3 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 |
|---|---|---|---|---|
| 17 Feb 2026 | GLM-5 technical report | 84.2 | GLM-5-Base, CLUEWSC (EM) (Table 11) | Capabilities |
| 8 Aug 2025 | GLM-4.5 technical report | 83.5 | GLM-4.5-Base, CLUEWSC (EM) (Table 2) | Capabilities |
| 20 Jan 2025 | Kimi k1.5 technical report | 91.7 | CLUEWSC (EM), short-CoT (Table 3) | Capabilities |
| 20 Jan 2025 | DeepSeek-R1 technical report | 92.8 | CLUEWSC (EM), DeepSeek-R1 (Table 8, same as final column of Table 3) | Capabilities |
| 26 Dec 2024 | DeepSeek-V3 technical report | 90.9 | CLUEWSC (EM), DeepSeek-V3 (Table 6) | Capabilities |
| 7 May 2024 | DeepSeek-V2 (MoE-236B, May 2024) technical report | 89.9 | CLUEWSC 5-shot, DeepSeek-V2 Chat (RL) (Table 3) | Capabilities |
| 5 Jan 2024 | DeepSeek LLM 67B technical report | 60 | CLUEWSC, DeepSeek 67B Chat (Table 6) | Capabilities |
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.
Inkling 88.7% - 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%