# Scale Labs SEAL Chinese (SEAL Chinese)

> Scale's SEAL leaderboard for Chinese: expert raters judge how well models answer about 1,000 prompts written by native Chinese speakers. BenchLM shows it as display-only reference data, excluded from rankings. Scale marked this board deprecated as of March 2025 and no longer adds models.

Canonical page: https://benchlm.ai/benchmarks/scale-chinese

- Category: [external](/external)
- Last updated: November 12, 2026 snapshot

## About SEAL Chinese

- Year: 2026
- Tasks: About 1,000 prompts per language
- Format: Elo-style rating
- Difficulty: External agent and model evaluation
- Paper: [Scale Labs leaderboard](https://labs.scale.com/leaderboard/chinese)

Single-turn prompts across general and culturally specific scenarios, rated by native-speaking experts. Scores are ratings on an Elo-style scale built from expert human comparisons, so only the gaps between rows carry meaning. BenchLM mirrors 21 published rows captured on November 12, 2026 snapshot.

SEAL Chinese is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (21 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [o1 (December 2024)](https://labs.scale.com/leaderboard/chinese) | OpenAI | 1165 |
| 2 | [o3-mini](/models/o3-mini) | OpenAI | 1156 |
| 3 | [Gemini 1.5 Pro (August 27, 2024)](https://labs.scale.com/leaderboard/chinese) | Google | 1120 |
| 4 | [o1-preview](/models/o1-preview) | OpenAI | 1120 |
| 5 | [Gemini 2.0 Pro (December 2024)](https://labs.scale.com/leaderboard/chinese) | Google | 1117 |
| 6 | [Gemini Pro Flash 2](https://labs.scale.com/leaderboard/chinese) | Google | 1115 |
| 7 | [Gemini 1.5 Pro (November 2024)](https://labs.scale.com/leaderboard/chinese) | Google | 1077 |
| 8 | [Gemini 2.0 Flash Thinking (January 2025)](https://labs.scale.com/leaderboard/chinese) | Google | 1060 |
| 9 | [DeepSeek-R1](/models/deepseek-r1) | DeepSeek | 1052 |
| 10 | [DeepSeek V3](/models/deepseek-v3) | DeepSeek | 1031 |
| 11 | [GPT-4o (August 2024)](https://labs.scale.com/leaderboard/chinese) | OpenAI | 1029 |
| 12 | [Gemini 1.5 Flash](/models/gemini-1-5-flash) | Google | 1015 |
| 13 | [Mistral Large 2](/models/mistral-large-2) | Mistral | 1006 |
| 14 | [DeepSeek V2 Chat](https://labs.scale.com/leaderboard/chinese) | Deepseek | 996 |
| 15 | [GPT-4 (November 2024)](https://labs.scale.com/leaderboard/chinese) | OpenAI | 985 |
| 16 | [Aya Expanse 32B](https://labs.scale.com/leaderboard/chinese) | Cohere | 967 |
| 17 | [Gemma 2 27B](/models/gemma-2-27b) | Google | 966 |
| 18 | [Claude 3.5 Sonnet (June 2024)](https://labs.scale.com/leaderboard/chinese) | Anthropic | 930 |
| 19 | [Qwen 2 72B Instruct](https://labs.scale.com/leaderboard/chinese) | Alibaba | 902 |
| 20 | [Llama 3.3 70B Instruct](https://labs.scale.com/leaderboard/chinese) | Meta | 883 |
| 21 | [Yi 1.5 34B Chat](https://labs.scale.com/leaderboard/chinese) | Unknown | 780 |

## FAQ

### What does SEAL Chinese measure?

Scale's SEAL leaderboard for Chinese: expert raters judge how well models answer about 1,000 prompts written by native Chinese speakers. BenchLM shows it as display-only reference data, excluded from rankings. Scale marked this board deprecated as of March 2025 and no longer adds models.

### Which model leads the published SEAL Chinese snapshot?

o1 (December 2024) currently leads the published SEAL Chinese snapshot with a score of 1165.

### How many models are evaluated on SEAL Chinese?

The November 12, 2026 snapshot contains 21 AI models.

### Does SEAL Chinese affect BenchLM's overall score?

Not directly. SEAL Chinese is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.