---
title: "中文排行榜"
description: "中文提问的排名。"
canonical: https://pirouter.ai/zh/rankings/chinese
last-updated: 2026-10-15
---

# 中文排行榜

> 中文提问的排名。

- 更新于: 2026-09-30
- 902,936 票 · 362 个模型
- 分数 · 越高越好

## 完整排名

| Rank | Rank spread | Model | Vendor | Score | Votes |
|---|---|---|---|---|---|
| 1 | 1–8 | Gemini 4 Argon (high) | Google | 1588 ±32 | 352 |
| 2 | 1–11 | [Claude Fable 5.1 \(max\)](https://pirouter.ai/zh/models/anthropic/claude-fable-5.1) | Anthropic | 1573 ±22 | 841 |
| 3 | 1–22 | [Claude Opus 5.5 \(high\)](https://pirouter.ai/zh/models/anthropic/claude-opus-5.5) | Anthropic | 1572 ±31 | 353 |
| 4 | 1–20 | Claude Fable 5 (high) | Anthropic | 1555 ±12 | 2812 |
| 5 | 1–21 | Claude Opus 4.6 (high) | Anthropic | 1551 ±9 | 5339 |
| 6 | 1–26 | [Gemini 3.7 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.7-flash) | Google | 1549 ±15 | 1646 |
| 7 | 1–22 | Claude Opus 5 (high) | Anthropic | 1548 ±10 | 4653 |
| 8 | 2–37 | Kimi K3 (max) | Moonshot | 1540 ±14 | 2028 |
| 9 | 2–42 | Qwen3.8 (max) | Alibaba | 1539 ±15 | 1672 |
| 10 | 3–37 | [GPT-5.6 Sol \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-sol) | OpenAI | 1538 ±12 | 2794 |
| 11 | 3–37 | Claude Opus 4.7 (high) | Anthropic | 1537 ±10 | 4142 |
| 12 | 3–42 | [Gemini 3.8 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.8-flash) | Google | 1535 ±14 | 1970 |
| 13 | 3–50 | [GLM 5.3 Flash](https://pirouter.ai/zh/models/z-ai/glm-5.3-flash) | Z.ai | 1532 ±16 | 1548 |
| 14 | 4–44 | [Gemini 3.6 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.6-flash) | Google | 1530 ±12 | 2687 |
| 15 | 3–58 | Muse Spark 1.3 (max) | Meta | 1530 ±20 | 902 |
| 16 | 5–44 | Muse Spark 1.1 | Meta | 1530 ±12 | 2888 |
| 17 | 3–53 | GLM 5.3 (max) | Z.ai | 1530 ±17 | 1292 |
| 18 | 7–43 | Gemini 3.1 Pro Preview | Google | 1528 ±7 | 8292 |
| 19 | 1–81 | Qwen3.7 Preview (max) | Alibaba | 1528 ±35 | 312 |
| 20 | 7–47 | Claude Opus 4.8 | Anthropic | 1527 ±10 | 4395 |
| 21 | 7–50 | [GPT-5.5](https://pirouter.ai/zh/models/openai/gpt-5.5) | OpenAI | 1526 ±10 | 4466 |
| 22 | 3–66 | Muse Spark | Meta | 1524 ±20 | 968 |
| 23 | 3–70 | [DeepSeek V4.1 Flash \(max\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4.1-flash) | DeepSeek | 1524 ±24 | 640 |
| 24 | 8–58 | [GLM 5.2 \(max\)](https://pirouter.ai/zh/models/z-ai/glm-5.2) | Z.ai | 1521 ±11 | 3232 |
| 25 | 2–92 | Muse Spark 1.2 (xhigh) | Meta | 1520 ±33 | 328 |
| 26 | 8–58 | Gemini 3 Pro | Google | 1520 ±10 | 3444 |
| 27 | 8–59 | [Gemini 3.5 Flash \(medium\)](https://pirouter.ai/zh/models/google/gemini-3.5-flash) | Google | 1517 ±11 | 3431 |
| 28 | 8–66 | Kimi K2.6 | Moonshot | 1517 ±13 | 2333 |
| 29 | 8–68 | GLM 5 | Z.ai | 1516 ±14 | 2003 |
| 30 | 8–70 | Qwen3.5 Preview (max) | Alibaba | 1516 ±15 | 1583 |
| 31 | 8–67 | Gemini 3 Flash | Google | 1516 ±12 | 2467 |
| 32 | 8–66 | [GPT-5.6 Terra \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-terra) | OpenAI | 1516 ±12 | 2907 |
| 33 | 11–62 | GLM 5.1 | Z.ai | 1515 ±10 | 4005 |
| 34 | 11–62 | MiMo V2.5 Pro | Xiaomi | 1515 ±10 | 4429 |
| 35 | 7–83 | [GPT-6 Astra \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-astra) | OpenAI | 1514 ±23 | 631 |
| 36 | 3–100 | MiMo V2.6 Pro | Xiaomi | 1511 ±33 | 299 |
| 37 | 8–84 | Hy3 | Tencent | 1511 ±21 | 759 |
| 38 | 11–74 | Grok 4.20 Beta1 | xAI | 1510 ±15 | 1731 |
| 39 | 14–70 | Claude Sonnet 4.6 | Anthropic | 1509 ±10 | 4519 |
| 40 | 17–71 | [GPT-5.4 \(high\)](https://pirouter.ai/zh/models/openai/gpt-5.4) | OpenAI | 1507 ±10 | 4143 |
| 41 | 8–100 | Grok 4.7 (xhigh) | xAI | 1507 ±26 | 484 |
| 42 | 13–84 | GPT-5.5 Instant | OpenAI | 1506 ±15 | 1699 |
| 43 | 17–79 | Qwen3.7 Plus | Alibaba | 1505 ±12 | 2671 |
| 44 | 8–101 | MiMo V2.6 Flash | Xiaomi | 1505 ±28 | 421 |
| 45 | 19–80 | Grok 4.5 | xAI | 1504 ±11 | 3040 |
| 46 | 16–84 | Ernie 5.1 | Baidu | 1504 ±13 | 2103 |
| 47 | 17–88 | Grok 4.6 (high) | xAI | 1503 ±14 | 1825 |
| 48 | 20–81 | Grok 4.20 Beta 0309 Reasoning | xAI | 1502 ±10 | 4243 |
| 49 | 20–84 | Claude Opus 4.5 20251101 High 32k | Anthropic | 1501 ±11 | 3041 |
| 50 | 20–84 | [DeepSeek V4 Pro 0813 \(high\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4-pro) | DeepSeek | 1501 ±11 | 3344 |
| 51 | 20–93 | GPT-5.2 Chat Latest | OpenAI | 1498 ±12 | 2393 |
| 52 | 20–95 | GPT-5.3 Chat Latest | OpenAI | 1498 ±13 | 2271 |
| 53 | 24–84 | Qwen3.5 397b A17b | Alibaba | 1498 ±8 | 6084 |
| 54 | 19–99 | Qwen3.8 27b | Alibaba | 1498 ±17 | 1282 |
| 55 | 23–91 | GPT-5.1 (high) | OpenAI | 1498 ±11 | 3423 |
| 56 | 16–103 | Ernie 5.0 Preview 1203 | Baidu | 1497 ±21 | 828 |
| 57 | 26–89 | Claude Opus 4.5 | Anthropic | 1496 ±9 | 5400 |
| 58 | 19–100 | GLM 4.7 | Z.ai | 1496 ±18 | 1020 |
| 59 | 26–92 | Dola Seed 2.0 Pro | ByteDance | 1496 ±9 | 5268 |
| 60 | 24–95 | [Claude Sonnet 5 \(high\)](https://pirouter.ai/zh/models/anthropic/claude-sonnet-5) | Anthropic | 1496 ±11 | 3488 |
| 61 | 26–93 | Kimi K2.5 (thinking) | Moonshot | 1495 ±9 | 4606 |
| 62 | 11–115 | Ernie 5.0 Preview 1022 | Baidu | 1495 ±30 | 381 |
| 63 | 30–93 | Gemini 3 Flash | Google | 1495 ±8 | 5989 |
| 64 | 11–116 | Qwen3.6 Preview (max) | Alibaba | 1493 ±31 | 354 |
| 65 | 16–111 | Gemma 4 26b A4b | Google | 1493 ±25 | 525 |
| 66 | 31–95 | Grok 4.1 | xAI | 1493 ±9 | 5285 |
| 67 | 31–98 | Grok 4.20 Multi Agent Beta 0309 | xAI | 1493 ±10 | 4045 |
| 68 | 36–100 | Gemini 2.5 Pro | Google | 1487 ±7 | 9256 |
| 69 | 34–108 | Ernie 5.0 0110 | Baidu | 1486 ±12 | 2546 |
| 70 | 36–108 | [Gemini 3.5 Flash Lite](https://pirouter.ai/zh/models/google/gemini-3.5-flash-lite) | Google | 1483 ±12 | 2726 |
| 71 | 36–109 | Qwen3.6 Plus | Alibaba | 1483 ±12 | 2664 |
| 72 | 24–121 | [GPT-6 Sol \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-sol) | OpenAI | 1482 ±24 | 625 |
| 73 | 35–112 | Longcat Flash Chat 2602 Exp | Meituan | 1482 ±14 | 1982 |
| 74 | 26–121 | [GPT-6 Luna \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-luna) | OpenAI | 1481 ±24 | 596 |
| 75 | 40–111 | [DeepSeek V4 Flash 0731 \(high\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4-flash) | DeepSeek | 1480 ±11 | 2987 |
| 76 | 48–108 | Claude Sonnet 4.5 20250929 High 32k | Anthropic | 1480 ±8 | 5870 |
| 77 | 36–116 | MiMo V2 Pro | Xiaomi | 1479 ±15 | 1559 |
| 78 | 41–113 | MiMo V2.5 | Xiaomi | 1479 ±12 | 2593 |
| 79 | 29–123 | Gemma 4 31b | Google | 1479 ±25 | 521 |
| 80 | 40–114 | Qwen3 Preview (max) | Alibaba | 1479 ±13 | 2065 |
| 81 | 50–112 | Grok 4.3 | xAI | 1477 ±10 | 4522 |
| 82 | 47–114 | Inkling | Thinky | 1477 ±12 | 2690 |
| 83 | 47–114 | GLM 4.6 | Z.ai | 1477 ±12 | 2694 |
| 84 | 47–114 | [GPT-5.6 Luna \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-luna) | OpenAI | 1477 ±12 | 2968 |
| 85 | 55–111 | Claude Sonnet 4.5 | Anthropic | 1477 ±8 | 5862 |
| 86 | 35–121 | MiMo V2 Omni | Xiaomi | 1477 ±19 | 1002 |
| 87 | 52–114 | Gemini 3.1 Flash Lite Preview | Google | 1477 ±10 | 4178 |
| 88 | 35–123 | GLM 5v Turbo | Z.ai | 1475 ±21 | 774 |
| 89 | 58–114 | Kimi K2 Thinking Turbo | Moonshot | 1474 ±9 | 4674 |
| 90 | 58–116 | [GPT-5.4 Mini \(high\)](https://pirouter.ai/zh/models/openai/gpt-5.4-mini) | OpenAI | 1474 ±10 | 3924 |
| 91 | 55–121 | Qwen3.5 27b | Alibaba | 1472 ±13 | 2049 |
| 92 | 59–120 | Minimax M3 | MiniMax | 1472 ±11 | 3634 |
| 93 | 47–125 | DeepSeek V3.1 (thinking) | DeepSeek | 1471 ±18 | 1137 |
| 94 | 66–121 | Chatgpt 4o Latest | OpenAI | 1468 ±8 | 5828 |
| 95 | 64–121 | Claude Opus 4.1 (thinking) | Anthropic | 1468 ±10 | 3833 |
| 96 | 66–121 | Qwen3 235b A22b Instruct 2507 | Alibaba | 1467 ±7 | 7250 |
| 97 | 37–142 | DeepSeek V3.2 Exp (thinking) | DeepSeek | 1467 ±26 | 504 |
| 98 | 49–138 | Nvidia Nemotron 3 Ultra 550b A55b Nvfp4 | NVIDIA | 1465 ±22 | 729 |
| 99 | 36–147 | Grok 4 Fast Chat | xAI | 1465 ±28 | 460 |
| 100 | 66–123 | GPT-5.2 (high) | OpenAI | 1464 ±10 | 3568 |

## FAQ

### 中文最好的 AI 模型是哪个？

截至 2026-09-30，中文榜第一为 Google Gemini 4 Argon（high），分数 1588，名次区间 1–8。

### 中文最好的开放权重模型是哪个？

Kimi K3（Moonshot），全榜第 8 名。

### 这个榜单排的是什么？

中文提问的排名。

### 多久更新一次？

有新模型发布或排名变化时更新；本榜最近一次更新为 2026-09-30。

数据：[LMArena leaderboard dataset](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)，按 [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) 许可使用；仅做筛选与推理档位合并，分数未改动。
查看原榜单: https://arena.ai/leaderboard/text/chinese