AIModelPort Hub

Ranking

Best Chinese LLM

Rank AI models for Chinese writing, Chinese QA, and localized workflows.

Answer Summary / AI 摘要

Updated 2026-09-02

Best Chinese LLM 使用 chinese 作为核心指标,对 233 个 AI 模型进行排序。English summary: this ranking compares AI models by chinese; the current top result is Qwen 3.5 Max.

  • Primary metric: chinese
  • Models included: 233
  • Top result: Qwen 3.5 Max

ModelPort API

测试榜单中的候选模型

看完 Best Chinese LLM 后,可以进入 ModelPort API 测试排名靠前的模型,并用自己的任务集比较质量、速度和成本。

Rank模型目录评分*
#1Qwen 3.5 Max

Alibaba Cloud

98/100
#2DeepSeek V4

DeepSeek

96/100
#3Qwen: Qwen3 Coder Next

Alibaba Cloud

90/100
#4Qwen: Qwen3 Max Thinking

Alibaba Cloud

90/100
#5Qwen: Qwen3 VL 30B A3B Thinking

Alibaba Cloud

90/100
#6Qwen: Qwen3 VL 32B Instruct

Alibaba Cloud

90/100
#7Qwen: Qwen3 VL 8B Instruct

Alibaba Cloud

90/100
#8Qwen: Qwen3 VL 8B Thinking

Alibaba Cloud

90/100
#9Qwen: Qwen3.5 397B A17B

Alibaba Cloud

90/100
#10Qwen: Qwen3.5 Plus 2026-02-15

Alibaba Cloud

90/100
#11Qwen: Qwen3.5 Plus 2026-04-20

Alibaba Cloud

90/100
#12Qwen: Qwen3.5-122B-A10B

Alibaba Cloud

90/100
#13Qwen: Qwen3.5-27B

Alibaba Cloud

90/100
#14Qwen: Qwen3.5-35B-A3B

Alibaba Cloud

90/100
#15Qwen: Qwen3.5-9B

Alibaba Cloud

90/100
#16Qwen: Qwen3.5-Flash

Alibaba Cloud

90/100
#17Qwen: Qwen3.6 27B

Alibaba Cloud

90/100
#18Qwen: Qwen3.6 35B A3B

Alibaba Cloud

90/100
#19Qwen: Qwen3.6 Flash

Alibaba Cloud

90/100
#20Qwen: Qwen3.6 Max Preview

Alibaba Cloud

90/100
#21Qwen: Qwen3.6 Plus

Alibaba Cloud

90/100
#22Qwen: Qwen3.7 Flash

Alibaba Cloud

90/100
#23Qwen: Qwen3.7 Max

Alibaba Cloud

90/100
#24Qwen: Qwen3.7 Plus

Alibaba Cloud

90/100
#25Qwen: Qwen3.8 2.4T A95B

Alibaba Cloud

90/100
#26Qwen: Qwen3.8 27B

Alibaba Cloud

90/100
#27Qwen: Qwen3.8 Flash

Alibaba Cloud

90/100
#28Qwen: Qwen3.8 Max

Alibaba Cloud

90/100
#29GPT-5.1

OpenAI

88/100
#30Gemini 3 Pro

Google

86/100
#31Claude 4.6 Sonnet

Anthropic

84/100
#32Mistral Large 3

Mistral AI

72/100
#33AionLabs: Aion-2.0

Other

72/100
#34AionLabs: Aion-3.0

Other

72/100
#35AionLabs: Aion-3.0-Mini

Other

72/100
#36AllenAI: Olmo 3 32B Think

Other

72/100
#37Amazon: Nova 2 Lite

Other

72/100
#38Amazon: Nova Premier 1.0

Other

72/100
#39Anthropic Claude Haiku Latest

Anthropic

72/100
#40Anthropic Claude Sonnet Latest

Anthropic

72/100
#41Anthropic: Claude Fable 5

Anthropic

72/100
#42Anthropic: Claude Fable 5.1

Anthropic

72/100
#43Anthropic: Claude Fable Latest

Anthropic

72/100
#44Anthropic: Claude Haiku 4.5

Anthropic

72/100
#45Anthropic: Claude Opus 4.5

Anthropic

72/100
#46Anthropic: Claude Opus 4.6

Anthropic

72/100
#47Anthropic: Claude Opus 4.7

Anthropic

72/100
#48Anthropic: Claude Opus 4.8

Anthropic

72/100
#49Anthropic: Claude Opus 4.8 (Fast)

Anthropic

72/100
#50Anthropic: Claude Opus Latest

Anthropic

72/100
#51Anthropic: Claude Sonnet 4.6

Anthropic

72/100
#52Anthropic: Claude Sonnet 5

Anthropic

72/100
#53Arcee AI: Trinity Large Thinking

Other

72/100
#54Arcee AI: Trinity Mini

Other

72/100
#55Auto Router (Beta)

Other

72/100
#56BAAI: bge-base-en-v1.5

Other

72/100
#57BAAI: bge-large-en-v1.5

Other

72/100
#58BAAI: bge-m3

Other

72/100
#59Body Builder (beta)

Other

72/100
#60ByteDance Seed: Seed 1.6

Other

72/100
#61ByteDance Seed: Seed 1.6 Flash

Other

72/100
#62ByteDance Seed: Seed 2.1 Turbo

Other

72/100
#63ByteDance Seed: Seed-2.0-Code

Other

72/100
#64ByteDance Seed: Seed-2.0-Lite

Other

72/100
#65ByteDance Seed: Seed-2.0-Mini

Other

72/100
#66Canopy Labs: Orpheus 3B

Other

72/100
#67Claude Opus 5

Anthropic

72/100
#68Claude Opus 5 (Fast)

Anthropic

72/100
#69Deep Cogito: Cogito v2.1 671B

Other

72/100
#70DeepSeek V4 Flash Latest

DeepSeek

72/100
#71DeepSeek: DeepSeek V3.2

DeepSeek

72/100
#72DeepSeek: DeepSeek V4 Flash 0423

DeepSeek

72/100
#73DeepSeek: DeepSeek V4 Flash 0731

DeepSeek

72/100
#74DeepSeek: DeepSeek V4 Flash Vision Exp

DeepSeek

72/100
#75DeepSeek: DeepSeek V4 Pro 0423

DeepSeek

72/100
#76DeepSeek: DeepSeek V4 Pro 0813

DeepSeek

72/100
#77EssentialAI: Rnj 1 Instruct

Other

72/100
#78Google Gemini Flash Latest

Google

72/100
#79Google Gemini Pro Latest

Google

72/100
#80Google: Gemini 3 Flash Preview

Google

72/100
#81Google: Gemini 3.1 Flash Lite

Google

72/100
#82Google: Gemini 3.1 Flash Lite Preview

Google

72/100
#83Google: Gemini 3.1 Pro Preview

Google

72/100
#84Google: Gemini 3.1 Pro Preview Custom Tools

Google

72/100
#85Google: Gemini 3.5 Flash

Google

72/100
#86Google: Gemini 3.5 Flash Lite

Google

72/100
#87Google: Gemini 3.6 Flash

Google

72/100
#88Google: Gemini 3.7 Flash

Google

72/100
#89Google: Gemma 4 26B A4B

Google

72/100
#90Google: Gemma 4 31B

Google

72/100
#91IBM: Granite 4.0 Micro

Other

72/100
#92IBM: Granite 4.1 8B

Other

72/100
#93IBM: Granite 4.2 8B

Other

72/100
#94Inception: Mercury 2

Other

72/100
#95Inception: Mercury 2.5 Preview

Other

72/100
#96Intfloat: E5-Base-v2

Other

72/100
#97Intfloat: E5-Large-v2

Other

72/100
#98Intfloat: Multilingual-E5-Large

Other

72/100
#99Kwaipilot: KAT-Coder-Air V2.5

Other

72/100
#100Kwaipilot: KAT-Coder-Pro V2

Other

72/100
#101Kwaipilot: KAT-Coder-Pro V2.5

Other

72/100
#102Ling-3.0-flash

Other

72/100
#103LiquidAI: LFM2-24B-A2B

Other

72/100
#104Meituan: LongCat 2.0

Other

72/100
#105Meta: Muse Glimmer 30B

Other

72/100
#106Meta: Muse Spark 1.1

Other

72/100
#107Meta: Muse Spark 1.2

Other

72/100
#108Meta: Muse Spark 1.2 Contributor

Other

72/100
#109Microsoft: Phi 4 Mini Instruct

Other

72/100
#110MiniMax: MiniMax M2

Other

72/100
#111MiniMax: MiniMax M2-her

Other

72/100
#112MiniMax: MiniMax M2.1

Other

72/100
#113MiniMax: MiniMax M2.5

Other

72/100
#114MiniMax: MiniMax M2.7

Other

72/100
#115MiniMax: MiniMax M3

Other

72/100
#116Mistral: Codestral Embed 2505

Mistral AI

72/100
#117Mistral: Devstral 2 2512

Mistral AI

72/100
#118Mistral: Ministral 3 14B 2512

Mistral AI

72/100
#119Mistral: Ministral 3 3B 2512

Mistral AI

72/100
#120Mistral: Ministral 3 8B 2512

Mistral AI

72/100
#121Mistral: Mistral Embed 2312

Mistral AI

72/100
#122Mistral: Mistral Large 3 2512

Mistral AI

72/100
#123Mistral: Mistral Medium 3.5

Mistral AI

72/100
#124Mistral: Mistral Small 4

Mistral AI

72/100
#125Mistral: Voxtral Small 24B 2507

Mistral AI

72/100
#126MoonshotAI Kimi Latest

Other

72/100
#127MoonshotAI: Kimi K2 Thinking

Other

72/100
#128MoonshotAI: Kimi K2.5

Other

72/100
#129MoonshotAI: Kimi K2.6

Other

72/100
#130MoonshotAI: Kimi K2.7 Code

Other

72/100
#131MoonshotAI: Kimi K3

Other

72/100
#132NVIDIA: Llama 3.3 Nemotron Super 49B V1.5

Other

72/100
#133NVIDIA: Nemotron 3 Nano 30B A3B

Other

72/100
#134NVIDIA: Nemotron 3 Super

Other

72/100
#135NVIDIA: Nemotron 3 Ultra

Other

72/100
#136NVIDIA: Nemotron 3.5 Lightning

Other

72/100
#137Nex AGI: DeepSeek V3.1 Nex N1

Other

72/100
#138Nex AGI: Nex-N2-Mini

Other

72/100
#139Nex AGI: Nex-N2-Pro

Other

72/100
#140OpenAI GPT Latest

OpenAI

72/100
#141OpenAI GPT Mini Latest

OpenAI

72/100
#142OpenAI: GPT Audio

OpenAI

72/100
#143OpenAI: GPT Audio Mini

OpenAI

72/100
#144OpenAI: GPT Chat Latest

OpenAI

72/100
#145OpenAI: GPT-5.1 Chat

OpenAI

72/100
#146OpenAI: GPT-5.1-Codex

OpenAI

72/100
#147OpenAI: GPT-5.1-Codex-Max

OpenAI

72/100
#148OpenAI: GPT-5.1-Codex-Mini

OpenAI

72/100
#149OpenAI: GPT-5.2

OpenAI

72/100
#150OpenAI: GPT-5.2 Chat

OpenAI

72/100
#151OpenAI: GPT-5.2-Codex

OpenAI

72/100
#152OpenAI: GPT-5.3 Chat

OpenAI

72/100
#153OpenAI: GPT-5.3-Codex

OpenAI

72/100
#154OpenAI: GPT-5.4

OpenAI

72/100
#155OpenAI: GPT-5.4 Mini

OpenAI

72/100
#156OpenAI: GPT-5.4 Nano

OpenAI

72/100
#157OpenAI: GPT-5.5

OpenAI

72/100
#158OpenAI: GPT-5.6 Luna

OpenAI

72/100
#159OpenAI: GPT-5.6 Luna Pro

OpenAI

72/100
#160OpenAI: GPT-5.6 Sol

OpenAI

72/100
#161OpenAI: GPT-5.6 Sol Pro

OpenAI

72/100
#162OpenAI: GPT-5.6 Terra

OpenAI

72/100
#163OpenAI: GPT-5.6 Terra Pro

OpenAI

72/100
#164OpenAI: gpt-oss-safeguard-20b

OpenAI

72/100
#165OpenAI: o3 Deep Research

OpenAI

72/100
#166OpenAI: o4 Mini Deep Research

OpenAI

72/100
#167OpenRouter: Fusion

Other

72/100
#168Pareto Code Router

Other

72/100
#169Perceptron: Perceptron Mk1

Other

72/100
#170Perplexity: Embed V1 0.6B

Other

72/100
#171Perplexity: Embed V1 4B

Other

72/100
#172Perplexity: Sonar Pro Search

Other

72/100
#173Poolside: Laguna M.1

Other

72/100
#174Poolside: Laguna S 2.1

Other

72/100
#175Poolside: Laguna XS 2.1

Other

72/100
#176Poolside: Laguna XS.2

Other

72/100
#177Prime Intellect: INTELLECT-3

Other

72/100
#178Reka Edge

Other

72/100
#179Relace: Relace Search

Other

72/100
#180Sakana: Fugu Ultra

Other

72/100
#181Sakana: Sakana Namazu

Other

72/100
#182Sentence Transformers: all-MiniLM-L12-v2

Other

72/100
#183Sentence Transformers: all-MiniLM-L6-v2

Other

72/100
#184Sentence Transformers: all-mpnet-base-v2

Other

72/100
#185Sentence Transformers: multi-qa-mpnet-base-dot-v1

Other

72/100
#186Sentence Transformers: paraphrase-MiniLM-L6-v2

Other

72/100
#187Sesame: CSM 1B

Other

72/100
#188SpaceXAI: Grok 4.20

Other

72/100
#189SpaceXAI: Grok 4.20 Multi-Agent

Other

72/100
#190SpaceXAI: Grok 4.3

Other

72/100
#191SpaceXAI: Grok 4.5

Other

72/100
#192SpaceXAI: Grok 4.6

Other

72/100
#193SpaceXAI: Grok Build 0.1

Other

72/100
#194StepFun: Step 3.5 Flash

Other

72/100
#195StepFun: Step 3.7 Flash

Other

72/100
#196Tencent: Hy-MT2-1.8B

Other

72/100
#197Tencent: Hy-MT2-30B-A3B

Other

72/100
#198Tencent: Hy-MT2-7B

Other

72/100
#199Tencent: Hy3

Other

72/100
#200Tencent: Hy3 preview

Other

72/100
#201Tencent: Hy4 preview

Other

72/100
#202Thenlper: GTE-Base

Other

72/100
#203Thenlper: GTE-Large

Other

72/100
#204Upstage: Solar Pro 3

Other

72/100
#205Upstage: Solar Pro 4

Other

72/100
#206VoyageAI by MongoDB: voyage-4

Other

72/100
#207VoyageAI by MongoDB: voyage-4-large

Other

72/100
#208VoyageAI by MongoDB: voyage-4-lite

Other

72/100
#209VoyageAI by MongoDB: voyage-code-4

Other

72/100
#210VoyageAI by MongoDB: voyage-multimodal-3.5

Other

72/100
#211Writer: Palmyra X5

Other

72/100
#212Xiaomi: MiMo-V2-Flash

Other

72/100
#213Xiaomi: MiMo-V2.5

Other

72/100
#214Xiaomi: MiMo-V2.5-Pro

Other

72/100
#215Z.ai: GLM 4.6V

Other

72/100
#216Z.ai: GLM 4.7

Other

72/100
#217Z.ai: GLM 4.7 Flash

Other

72/100
#218Z.ai: GLM 5

Other

72/100
#219Z.ai: GLM 5 Turbo

Other

72/100
#220Z.ai: GLM 5.1

Other

72/100
#221Z.ai: GLM 5.2

Other

72/100
#222Z.ai: GLM 5.3

Other

72/100
#223Z.ai: GLM 5.3 Flash

Other

72/100
#224Z.ai: GLM 5V Turbo

Other

72/100
#225Z.ai: GLM Flash Latest

Other

72/100
#226Z.ai: GLM Latest

Other

72/100
#227Zyphra: Zonos v0.1 Hybrid

Other

72/100
#228Zyphra: Zonos v0.1 Transformer

Other

72/100
#229hexgrad: Kokoro 82M

Other

72/100
#230inclusionAI: Ling-2.6-1T

Other

72/100
#231inclusionAI: Ling-2.6-flash

Other

72/100
#232inclusionAI: Ring-2.6-1T

Other

72/100
#233xAI: Grok Latest

Other

72/100

* 目录评分用于候选模型排序,基于编辑判断与目录能力信号,不代表独立 benchmark 成绩。生产选型请使用自己的任务集复测。

Facts Table / Source Facts

字段含义、来源优先级与更新时间规则见数据与排名方法

Machine-readable facts for AI search engines and answer engines.
FieldValueNote
RankingBest Chinese LLMCatalog data
Primary metricchineseCatalog data
Models included233Catalog data
MethodologySort by editorial catalog signals, then review price, context window, vision support, and function calling. The score is not an independent benchmark result.Catalog data
Updated2026-09-02Catalog data

排名方法

Sort by editorial catalog signals, then review price, context window, vision support, and function calling. The score is not an independent benchmark result.

FAQ

Best Chinese LLM 排名如何计算?+

排名由结构化目录数据和编辑判断生成,结合价格、能力信号、上下文长度和关键 API 能力;目录评分不是独立 benchmark 成绩。

这个排行榜适合直接做采购决策吗?+

它适合做候选列表,生产采购前仍需要用自己的任务样本做评估。

引用格式 / Citation Format

AI 搜索或研究型回答可以引用下面的稳定格式。

Plain text

ModelPort Hub. "Best Chinese LLM: AI Model Ranking." Updated 2026-09-02. https://modelporthub.com/rankings/best-chinese-llm

Markdown

[Best Chinese LLM: AI Model Ranking](https://modelporthub.com/rankings/best-chinese-llm) — ModelPort Hub, updated 2026-09-02.