---
title: "长文本排行榜"
description: "长输入提问的排名。"
canonical: https://pirouter.ai/zh/rankings/longctx
last-updated: 2026-10-15
---

# 长文本排行榜

> 长输入提问的排名。

- 更新于: 2026-09-30
- 2,821,105 票 · 363 个模型
- 分数 · 越高越好

## 完整排名

| Rank | Rank spread | Model | Vendor | Score | Votes |
|---|---|---|---|---|---|
| 1 | 1–2 | Gemini 4 Argon (high) | Google | 1544 ±14 | 2079 |
| 2 | 1–7 | [Claude Opus 5.5 \(high\)](https://pirouter.ai/zh/models/anthropic/claude-opus-5.5) | Anthropic | 1528 ±15 | 1708 |
| 3 | 2–6 | Claude Opus 4.6 (high) | Anthropic | 1524 ±5 | 32696 |
| 4 | 2–6 | Claude Fable 5 (high) | Anthropic | 1523 ±6 | 17723 |
| 5 | 2–9 | Claude Opus 4.7 (high) | Anthropic | 1515 ±5 | 29893 |
| 6 | 2–17 | [Claude Fable 5.1 \(max\)](https://pirouter.ai/zh/models/anthropic/claude-fable-5.1) | Anthropic | 1511 ±9 | 4705 |
| 7 | 4–18 | [Gemini 3.8 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.8-flash) | Google | 1508 ±7 | 11720 |
| 8 | 5–18 | Claude Opus 5 (high) | Anthropic | 1505 ±5 | 27765 |
| 9 | 5–21 | Kimi K3 (max) | Moonshot | 1504 ±7 | 12240 |
| 10 | 6–22 | Claude Opus 4.8 (high) | Anthropic | 1502 ±5 | 30035 |
| 11 | 6–24 | [Gemini 3.7 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.7-flash) | Google | 1501 ±7 | 9950 |
| 12 | 6–28 | Muse Spark 1.3 (max) | Meta | 1500 ±9 | 5391 |
| 13 | 6–24 | Gemini 3.1 Pro Preview | Google | 1499 ±4 | 53616 |
| 14 | 6–27 | [GPT-5.6 Sol \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-sol) | OpenAI | 1498 ±6 | 16784 |
| 15 | 7–32 | Claude Opus 4.5 20251101 High 32k | Anthropic | 1495 ±7 | 9830 |
| 16 | 6–47 | MiMo V2.6 Pro | Xiaomi | 1495 ±15 | 1661 |
| 17 | 6–47 | Muse Spark 1.2 (xhigh) | Meta | 1494 ±14 | 1767 |
| 18 | 9–33 | Claude Opus 4.5 | Anthropic | 1493 ±5 | 24253 |
| 19 | 9–34 | Claude Sonnet 4.6 | Anthropic | 1492 ±5 | 31045 |
| 20 | 6–50 | Qwen3.7 Preview (max) | Alibaba | 1492 ±15 | 1712 |
| 21 | 10–42 | Gemini 3 Pro | Google | 1491 ±6 | 11136 |
| 22 | 9–45 | Qwen3.8 (max) | Alibaba | 1490 ±7 | 10284 |
| 23 | 11–47 | GLM 5.3 (max) | Z.ai | 1489 ±8 | 8106 |
| 24 | 13–46 | [GPT-5.5 \(high\)](https://pirouter.ai/zh/models/openai/gpt-5.5) | OpenAI | 1487 ±5 | 32011 |
| 25 | 14–47 | MiMo V2.5 Pro | Xiaomi | 1487 ±5 | 31419 |
| 26 | 13–47 | [GLM 5.2 \(max\)](https://pirouter.ai/zh/models/z-ai/glm-5.2) | Z.ai | 1487 ±6 | 20853 |
| 27 | 13–47 | [Gemini 3.6 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.6-flash) | Google | 1487 ±6 | 16981 |
| 28 | 15–47 | Claude Sonnet 4.5 20250929 High 32k | Anthropic | 1486 ±5 | 26047 |
| 29 | 11–51 | [GPT-6 Astra \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-astra) | OpenAI | 1486 ±10 | 3656 |
| 30 | 15–47 | Claude Opus 4.1 (thinking) | Anthropic | 1485 ±6 | 11441 |
| 31 | 16–50 | Muse Spark 1.1 | Meta | 1482 ±6 | 17072 |
| 32 | 18–49 | Claude Sonnet 4.5 | Anthropic | 1482 ±5 | 26410 |
| 33 | 17–50 | [Claude Sonnet 5 \(high\)](https://pirouter.ai/zh/models/anthropic/claude-sonnet-5) | Anthropic | 1482 ±6 | 21121 |
| 34 | 18–50 | [Gemini 3.5 Flash \(high\)](https://pirouter.ai/zh/models/google/gemini-3.5-flash) | Google | 1482 ±5 | 22623 |
| 35 | 15–53 | Qwen3.5 Preview (max) | Alibaba | 1481 ±7 | 8958 |
| 36 | 18–51 | [GPT-5.4 \(high\)](https://pirouter.ai/zh/models/openai/gpt-5.4) | OpenAI | 1481 ±5 | 28408 |
| 37 | 18–52 | Grok 4.5 | xAI | 1481 ±6 | 19059 |
| 38 | 18–57 | [GLM 5.3 Flash](https://pirouter.ai/zh/models/z-ai/glm-5.3-flash) | Z.ai | 1479 ±7 | 9962 |
| 39 | 19–54 | GLM 5.1 | Z.ai | 1479 ±5 | 26298 |
| 40 | 15–62 | [DeepSeek V4.1 Flash \(max\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4.1-flash) | DeepSeek | 1478 ±10 | 3644 |
| 41 | 19–57 | Kimi K2.6 | Moonshot | 1478 ±6 | 16940 |
| 42 | 18–62 | Muse Spark | Meta | 1477 ±9 | 5602 |
| 43 | 20–62 | Gemini 3 Flash | Google | 1476 ±7 | 8450 |
| 44 | 21–62 | GPT-5.5 Instant | OpenAI | 1475 ±7 | 12539 |
| 45 | 18–65 | [DeepSeek V4 Pro 0813 \(high\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4-pro) | DeepSeek | 1475 ±9 | 4582 |
| 46 | 18–72 | Qwen3.6 Preview (max) | Alibaba | 1473 ±13 | 2294 |
| 47 | 29–64 | GPT-5.2 Chat Latest | OpenAI | 1473 ±6 | 13953 |
| 48 | 29–65 | Grok 4.6 (high) | xAI | 1471 ±7 | 10868 |
| 49 | 19–81 | Gemma 4 31b | Google | 1470 ±13 | 1799 |
| 50 | 37–69 | GLM 5 | Z.ai | 1468 ±6 | 11109 |
| 51 | 39–70 | [GPT-5.6 Terra \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-terra) | OpenAI | 1467 ±6 | 17440 |
| 52 | 30–82 | [GPT-6 Sol \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-sol) | OpenAI | 1466 ±11 | 3182 |
| 53 | 38–73 | Claude Opus 4 (thinking) | Anthropic | 1466 ±8 | 6923 |
| 54 | 39–73 | MiMo V2 Pro | Xiaomi | 1466 ±7 | 9979 |
| 55 | 41–70 | Grok 4.20 Beta 0309 Reasoning | xAI | 1466 ±5 | 29104 |
| 56 | 36–81 | Hy3 | Tencent | 1466 ±9 | 4727 |
| 57 | 39–73 | Grok 4.20 Beta1 | xAI | 1466 ±7 | 10448 |
| 58 | 41–73 | Qwen3.7 Plus | Alibaba | 1465 ±6 | 19490 |
| 59 | 41–73 | Ernie 5.1 | Baidu | 1465 ±6 | 16703 |
| 60 | 34–90 | MiMo V2.6 Flash | Xiaomi | 1463 ±13 | 2252 |
| 61 | 45–82 | [Gemini 3.5 Flash Lite](https://pirouter.ai/zh/models/google/gemini-3.5-flash-lite) | Google | 1462 ±6 | 16906 |
| 62 | 46–80 | Gemini 3 Flash | Google | 1462 ±5 | 36469 |
| 63 | 41–90 | [GPT-6 Luna \(max\)](https://pirouter.ai/zh/models/openai/gpt-6-luna) | OpenAI | 1460 ±11 | 3301 |
| 64 | 45–84 | GPT-5.1 (high) | OpenAI | 1460 ±6 | 10924 |
| 65 | 48–84 | Kimi K2.5 (thinking) | Moonshot | 1458 ±5 | 29804 |
| 66 | 48–84 | Grok 4.20 Multi Agent Beta 0309 | xAI | 1458 ±5 | 27931 |
| 67 | 48–90 | GPT-5.3 Chat Latest | OpenAI | 1457 ±6 | 13425 |
| 68 | 51–84 | Gemini 2.5 Pro | Google | 1456 ±4 | 33911 |
| 69 | 51–91 | Qwen3.6 Plus | Alibaba | 1455 ±6 | 21145 |
| 70 | 57–91 | Qwen3.5 397b A17b | Alibaba | 1453 ±4 | 38870 |
| 71 | 57–94 | [GPT-5.6 Luna \(xhigh\)](https://pirouter.ai/zh/models/openai/gpt-5.6-luna) | OpenAI | 1453 ±6 | 17820 |
| 72 | 57–94 | Dola Seed 2.0 Pro | ByteDance | 1452 ±5 | 32620 |
| 73 | 57–96 | MiMo V2.5 | Xiaomi | 1452 ±6 | 21290 |
| 74 | 58–96 | Minimax M3 | MiniMax | 1452 ±5 | 26432 |
| 75 | 49–102 | GLM 4.7 | Z.ai | 1451 ±10 | 3265 |
| 76 | 60–97 | Grok 4.1 | xAI | 1451 ±5 | 21142 |
| 77 | 48–106 | Grok 4.7 (xhigh) | xAI | 1451 ±13 | 2342 |
| 78 | 57–102 | MiMo V2 Omni | Xiaomi | 1450 ±8 | 9193 |
| 79 | 62–99 | [DeepSeek V4 Flash 0731 \(high\)](https://pirouter.ai/zh/models/deepseek/deepseek-v4-flash) | DeepSeek | 1449 ±6 | 23432 |
| 80 | 41–122 | DeepSeek V3.1 Terminus (thinking) | DeepSeek | 1447 ±22 | 672 |
| 81 | 66–102 | [GPT-5.4 Mini \(high\)](https://pirouter.ai/zh/models/openai/gpt-5.4-mini) | OpenAI | 1447 ±5 | 27712 |
| 82 | 57–113 | DeepSeek V3.1 (thinking) | DeepSeek | 1447 ±12 | 2367 |
| 83 | 52–118 | Gemma 4 26b A4b | Google | 1446 ±14 | 1717 |
| 84 | 66–108 | Qwen3.8 27b | Alibaba | 1445 ±8 | 7982 |
| 85 | 57–118 | GPT-4.5 Preview 2025 02 27 | OpenAI | 1443 ±14 | 1796 |
| 86 | 66–113 | Qwen3 Preview (max) | Alibaba | 1443 ±8 | 6126 |
| 87 | 71–106 | Grok 4.3 | xAI | 1443 ±5 | 33214 |
| 88 | 66–113 | Claude Sonnet 4 (thinking) | Anthropic | 1443 ±8 | 6687 |
| 89 | 71–113 | DeepSeek V3.2 (thinking) | DeepSeek | 1442 ±6 | 11730 |
| 90 | 62–118 | Kimi K2.5 Instant | Moonshot | 1442 ±12 | 2387 |
| 91 | 66–118 | GLM 5v Turbo | Z.ai | 1442 ±9 | 4889 |
| 92 | 66–118 | DeepSeek V3.2 Exp | DeepSeek | 1442 ±11 | 3045 |
| 93 | 75–111 | GPT-5.2 | OpenAI | 1441 ±5 | 34011 |
| 94 | 73–114 | Ernie 5.0 0110 | Baidu | 1440 ±6 | 11915 |
| 95 | 73–114 | Inkling | Thinky | 1440 ±6 | 16418 |
| 96 | 76–113 | Chatgpt 4o Latest | OpenAI | 1440 ±5 | 17699 |
| 97 | 76–118 | Longcat Flash Chat 2602 Exp | Meituan | 1437 ±7 | 11436 |
| 98 | 71–122 | Ernie 5.0 Preview 1203 | Baidu | 1436 ±12 | 2569 |
| 99 | 80–118 | Claude Haiku 4.5 | Anthropic | 1436 ±4 | 53867 |
| 100 | 77–120 | GPT-5 Chat | OpenAI | 1436 ±8 | 6795 |

## FAQ

### 长文本最好的 AI 模型是哪个？

截至 2026-09-30，长文本榜第一为 Google Gemini 4 Argon（high），分数 1544，名次区间 1–2。

### 长文本最好的开放权重模型是哪个？

Kimi K3（Moonshot），全榜第 9 名。

### 这个榜单排的是什么？

长输入提问的排名。

### 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/longer-query