# Chinese

> Source: Benchize (https://benchize.com) — https://benchize.com/benchmarks/scale%3Achinese. When citing or reusing this content, credit "Benchize" and link to https://benchize.com/benchmarks/scale%3Achinese.

The Scale AI Multilingual Prompts Dataset, composed of 1,000 prompts per language, is tailored to enhance models’ interaction capabilities across multiple languages. This dataset specifically aims to refine chatbots' proficiency in engaging with Chinese users, reflecting complexity of Chinese communication. This dataset introduces single-turn prompts across a diversified range of scenarios, aiming to evaluate and improve models’ responses in both general and culturally nuanced conversations.

Canonical URL: https://benchize.com/benchmarks/scale%3Achinese
Last updated: Wed, 21 Oct 2026 18:52:20 GMT

Reproduction and quotation must name Benchize, link to the canonical HTML URL, and retain individual source citations.

The Scale AI Multilingual Prompts Dataset, composed of 1,000 prompts per language, is tailored to enhance models’ interaction capabilities across multiple languages. This dataset specifically aims to refine chatbots' proficiency in engaging with Chinese users, reflecting complexity of Chinese communication. This dataset introduces single-turn prompts across a diversified range of scenarios, aiming to evaluate and improve models’ responses in both general and culturally nuanced conversations.

Language & multilingual · Models tested: 5 · Score publishers: [scale](https://labs.scale.com/leaderboard/chinese)

## Model results

| # | Model | Score | ± |  |
| --- | --- | --- | --- | --- |
| 1 | [O3 Mini](https://benchize.com/models/o3-mini.md) | 1,156source score | CI 1,124–1,188 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 1156 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration o3-mini sourceModelVersion o3-mini-2025-01-31 (high) thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 1,124–1,188 contaminationWarning Not reported rank 2 reportedConfidenceHalfWidth 32 ### Source [Original source](https://labs.scale.com/leaderboard/chinese) Source scale Original model label o3-mini Observed 2026-10-21T16:35:42.721Z Evidence SHA-256 `1deb2890a384b5b5eac4a0fb9aae797426e2c4cff6054f49e15508dfa6236956` Source locator `/rows/1` |  |  |  |  |
| 2 | [DeepSeek R1](https://benchize.com/models/deepseek-r1.md) | 1,052source score | CI 1,020–1,084 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 1052 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration DeepSeek R1 sourceModelVersion Not reported thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 1,020–1,084 contaminationWarning Not reported rank 9 reportedConfidenceHalfWidth 32 ### Source [Original source](https://labs.scale.com/leaderboard/chinese) Source scale Original model label DeepSeek R1 Observed 2026-10-21T16:35:42.721Z Evidence SHA-256 `1deb2890a384b5b5eac4a0fb9aae797426e2c4cff6054f49e15508dfa6236956` Source locator `/rows/8` |  |  |  |  |
| 3 | [Gemini 1.5 Flash](https://benchize.com/models/gemini-1-5-flash.md) | 1,015source score | CI 962–1,068 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 1015 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration Gemini 1.5 Flash sourceModelVersion gemini-1.5-flash-preview-0514 thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 962–1,068 contaminationWarning Not reported rank 12 reportedConfidenceHalfWidth 53 ### Source [Original source](https://labs.scale.com/leaderboard/chinese) Source scale Original model label Gemini 1.5 Flash Observed 2026-10-21T16:35:42.721Z Evidence SHA-256 `1deb2890a384b5b5eac4a0fb9aae797426e2c4cff6054f49e15508dfa6236956` Source locator `/rows/11` |  |  |  |  |
| 4 | [Gemma 2 27B](https://benchize.com/models/gemma-2-27b-it.md) | 966source score | CI 937–995 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 966 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration Gemma 2 27B sourceModelVersion gemma-2-27b thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 937–995 contaminationWarning Not reported rank 17 reportedConfidenceHalfWidth 29 ### Source [Original source](https://labs.scale.com/leaderboard/chinese) Source scale Original model label Gemma 2 27B Observed 2026-10-21T16:35:42.721Z Evidence SHA-256 `1deb2890a384b5b5eac4a0fb9aae797426e2c4cff6054f49e15508dfa6236956` Source locator `/rows/16` |  |  |  |  |
| 5 | [Llama 3.3 70B Instruct](https://benchize.com/models/llama-3-3-70b-instruct.md) | 883source score | CI 850–916 |  |
| ### Original source values \| Metric \| Value \| Unit \| Scope \| Direction \| \| --- \| --- \| --- \| --- \| --- \| \| `score` \| 883 \| source_score \| reported \| Higher is better \| ### Configuration Reasoning effort unknown Model configuration Llama 3.3 70B Instruct sourceModelVersion llama-3-3-70b-instruct thinkingEnabled Not reported ### Reported statistics Uncertainty Confidence interval: 850–916 contaminationWarning Not reported rank 20 reportedConfidenceHalfWidth 33 ### Source [Original source](https://labs.scale.com/leaderboard/chinese) Source scale Original model label Llama 3.3 70B Instruct Observed 2026-10-21T16:35:42.721Z Evidence SHA-256 `1deb2890a384b5b5eac4a0fb9aae797426e2c4cff6054f49e15508dfa6236956` Source locator `/rows/19` |  |  |  |  |

---
Cite as: Chinese — Benchize, https://benchize.com/benchmarks/scale%3Achinese (retrieved 2026-10-21)