模型榜
数据来源:AIHOT
只用第三方评测机构自己给模型实测的成绩,厂商自己公布的分数一概不计;各项成绩折算成同一种分数后按领域加权:编程 38%、科研 19%、专业办公 19%、知识问答 14%、真人盲选 10%。AIHOT 评分把各家难度不同的评测折算成同一种分数:50 分是参照模型的平均水平,每 15 分约是一个标准差;两个模型相差 3 分以上,先后基本不会变。评分不是正确率。
54 项评测 · 16 家机构 · 更新 2026-11-18 14:08 北京时间
国产由厂商推断,开源依据模型页的权重链接;暂不排名的模型缺少权重标记,开源筛选时不展示。价格单位:¥ / 百万 Token。
| 名次 | 模型 | 上线日期 | 评测 | 缓存价 | 输入价 | 输出价 | 评分 |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.2Anthropic | 2026-11-09 | 40 项评测 | ¥1.33 | ¥53.33 | ¥266.64 | 79.9 |
| 2 | Gemini 4 CarbonGoogle | 2026-10-28 | 31 项评测 | ¥1.33 | ¥26.66 | ¥133.32 | 77.0 |
| 3 | gpt-6.2-astraOpenAI | 2026-11-17 | 29 项评测 | ¥2.67 | ¥53.33 | ¥266.64 | 77.0 |
| 4 | Claude Opus 5.5Anthropic | 2026-09-22 | 57 项评测 | ¥1.33 | ¥26.66 | ¥133.32 | 74.1 |
| 5 | gpt-6.2-solOpenAI | 2026-11-17 | 25 项评测 | ¥0.67 | ¥13.33 | ¥66.66 | 73.5 |
| 6 | Gemini 4 HydrogenGoogle | 2026-11-12 | 21 项评测 | ¥0.5 | ¥10 | ¥50 | 72.7 |
| 7 | Claude Fable 5.1Anthropic | 2026-09-01 | 50 项评测 | ¥1.67 | ¥66.66 | ¥333.31 | 72.3 |
| 8 | GPT-6 AstraOpenAI | 2026-09-03 | 57 项评测 | ¥6.67 | ¥66.66 | ¥333.31 | 71.8 |
| 9 | Gemini 4 ArgonGoogle | 2026-09-30 | 51 项评测 | ¥1.33 | ¥26.66 | ¥133.32 | 71.7 |
| 10 | Claude Sonnet 5.5Anthropic | 2026-09-28 | 57 项评测 | ¥0.67 | ¥13.33 | ¥66.66 | 71.5 |
| 11 | GPT-6.1 SolOpenAI | 2026-09-29 | 56 项评测 | ¥0.67 | ¥13.33 | ¥66.66 | 70.0 |
| 12 | Claude Opus 5Anthropic | 2026-07-24 | 47 项评测 | ¥3.33 | ¥33.33 | ¥166.65 | 69.2 |
| 13 | GPT-6.1 TerraOpenAI | 2026-10-26 | 32 项评测 | ¥0.33 | ¥6.67 | ¥33.33 | 69.0 |
| 14 | Claude Fable 5Anthropic | 2026-06-09 | 38 项评测 | ¥6.67 | ¥66.66 | ¥333.31 | 68.7 |
| 15 | Qwen4 MaxAlibaba | 2026-11-16 | 24 项评测 | ¥0.5 | ¥10 | ¥30 | 68.7 |
| 16 | Kimi K3.1 ProMoonshot AI | 2026-10-23 | 38 项评测 | ¥1.3 | ¥13 | ¥68 | 67.8 |
| 17 | Muse WonderMeta | 2026-10-12 | 43 项评测 | ¥1.33 | ¥26.66 | ¥133.32 | 67.3 |
| 18 | GPT-5.6 SolOpenAI | 2026-07-09 | 43 项评测 | ¥3.33 | ¥33.33 | ¥199.98 | 66.7 |
| 19 | Hy4Tencent | 2026-11-04 | 24 项评测 | ¥0.3 | ¥6 | ¥18 | 66.7 |
| 20 | Muse EmberMeta | 2026-11-09 | 22 项评测 | ¥0.0267 | ¥0.67 | ¥3.33 | 66.0 |
| 21 | DeepSeek V4.1 Flash 1017DeepSeek | 2026-10-17 | 39 项评测 | ¥0.02 | ¥1 | ¥4 | 65.8 |
| 22 | GPT-6 SolOpenAI | 2026-09-22 | 47 项评测 | ¥1.33 | ¥13.33 | ¥66.66 | 65.5 |
| 23 | Muse Spark 1.3Meta | 2026-09-02 | 42 项评测 | ¥1 | ¥8.33 | ¥28.33 | 64.2 |
| 24 | Gemini 3.8 FlashGoogle | 2026-09-02 | 55 项评测 | ¥1 | ¥10 | ¥50 | 63.6 |
| 25 | Grok 4.7xAI | 2026-09-21 | 57 项评测 | ¥3.33 | ¥13.33 | ¥40 | 63.5 |
| 26 | GLM-5.3Z.ai | 2026-08-18 | 53 项评测 | ¥2 | ¥8 | ¥28 | 63.1 |
| 27 | Kimi K3Moonshot AI | 2026-07-16 | 55 项评测 | ¥2 | ¥20 | ¥100 | 62.9 |
| 28 | Qwen3.8 Max 0902Alibaba | 2026-09-01 | 30 项评测 | ¥1.5 | ¥12 | ¥36 | 62.9 |
| 29 | GPT-5.5OpenAI | 2026-04-23 | 33 项评测 | ¥3.33 | ¥33.33 | ¥199.98 | 62.8 |
| 30 | MiMo-V2.6-ProXiaomi | 2026-09-21 | 36 项评测 | ¥0.025 | ¥3 | ¥6 | 62.7 |
| 31 | Gemini 3.7 FlashGoogle | 2026-08-13 | 38 项评测 | ¥1 | ¥10 | ¥50 | 62.5 |
| 32 | Claude Haiku 5.5Anthropic | 2026-10-07 | 54 项评测 | ¥0.0667 | ¥0.67 | ¥3.33 | 62.3 |
| 33 | DeepSeek V4.1 FlashDeepSeek | 2026-09-10 | 47 项评测 | ¥0.02 | ¥1 | ¥4 | 62.3 |
| 34 | Grok 4.6xAI | 2026-08-12 | 45 项评测 | ¥3.33 | ¥13.33 | ¥40 | 62.2 |
| 35 | Inkling 2 SmallThinking Machines Lab | 2026-11-05 | 30 项评测 | ¥0.2 | ¥2 | ¥6.67 | 62.2 |
| 36 | Claude Opus 4.8Anthropic | 2026-05-28 | 37 项评测 | ¥3.33 | ¥33.33 | ¥166.65 | 62.2 |
| 37 | Hy 4 PreviewTencent | 2026-08-28 | 19 项评测 | ¥0.3 | ¥6 | ¥18 | 62.2 |
| 38 | Qwen3.8 MaxAlibaba | 2026-07-19 | 42 项评测 | ¥1.5 | ¥12 | ¥36 | 62.1 |
| 39 | GPT-5.6 TerraOpenAI | 2026-07-09 | 41 项评测 | ¥1.33 | ¥13.33 | ¥79.99 | 62.0 |
| 40 | Claude Opus 4.7Anthropic | 2026-04-16 | 29 项评测 | ¥3.33 | ¥33.33 | ¥166.65 | 60.7 |
| 41 | Step 5 PreviewStepFun | 2026-09-18 | 32 项评测 | ¥0.7 | ¥7 | ¥20 | 60.5 |
| 42 | Muse Spark 1.2Meta | 2026-08-05 | 32 项评测 | ¥1 | ¥8.33 | ¥28.33 | 60.0 |
| 43 | MiniMax-M3.1-FlashMiniMax | 2026-10-19 | 26 项评测 | ¥0.1 | ¥1 | ¥4 | 59.8 |
| 44 | Qwen3.8 Flash NextAlibaba | 2026-08-26 | 20 项评测 | ¥0.1 | ¥0.8 | ¥2.7 | 59.6 |
| 45 | DeepSeek V4 Pro 0813DeepSeek | 2026-08-13 | 46 项评测 | ¥0.15 | ¥4.5 | ¥13.5 | 59.6 |
| 46 | Claude Sonnet 5Anthropic | 2026-06-30 | 37 项评测 | ¥1.33 | ¥13.33 | ¥66.66 | 59.3 |
| 47 | MiMo V 2.6 FlashXiaomi | 2026-09-21 | 30 项评测 | ¥0.02 | ¥1 | ¥2 | 59.0 |
| 48 | Grok 4.5xAI | 2026-07-08 | 36 项评测 | ¥2 | ¥13.33 | ¥40 | 58.9 |
| 49 | GPT-5.4OpenAI | 2026-03-05 | 17 项评测 | ¥1.67 | ¥16.67 | ¥99.99 | 58.8 |
| 50 | GLM-5.3 FlashZ.ai | 2026-08-26 | 49 项评测 | ¥0.23 | ¥0.8 | ¥2.8 | 58.6 |
一般用量:缓存命中:输入:输出 = 7:2:1。 2026-11-17 · $1 = ¥6.6661
暂不排名 · 5
Muse Wonder FullMeta
70.1本轮未收录官方 API 价格或公开权重
ember-1Fireworks AI
63.3评测数量还不够
61.1
Ling 3.1 FlashInclusionAI
60.1评测数量还不够
Qwen3.8 2.4 T A 95 BAlibaba
60评测数量还不够
全部模型 · 45
| 名次 | 模型 | 上线日期 | 评测 | 缓存价 | 输入价 | 输出价 | 评分 |
|---|---|---|---|---|---|---|---|
| 51 | GPT-6 LunaOpenAI | 2026-09-22 | 53 项评测 | ¥0.0667 | ¥0.67 | ¥3.33 | 58.6 |
| 52 | Claude Opus 4.6Anthropic | 2026-02-05 | 23 项评测 | ¥3.33 | ¥33.33 | ¥166.65 | 58.4 |
| 53 | Muse Spark 1.1Meta | 2026-07-09 | 25 项评测 | ¥1 | ¥8.33 | ¥28.33 | 58.3 |
| 54 | DeepSeek V4 Flash 0731DeepSeek | 2026-07-31 | 31 项评测 | ¥0.05 | ¥1.5 | ¥4.5 | 58.0 |
| 55 | Gemini 3.6 FlashGoogle | 2026-07-21 | 35 项评测 | ¥1 | ¥10 | ¥50 | 57.1 |
| 56 | GPT-5.6 LunaOpenAI | 2026-07-09 | 44 项评测 | ¥0.13 | ¥1.33 | ¥8 | 57.1 |
| 57 | Muse Spark OpenMeta | 2026-10-15 | 27 项评测 | — | — | — | 57.0 |
| 58 | Gemini 3.5 FlashGoogle | 2026-05-19 | 34 项评测 | ¥1 | ¥10 | ¥59.99 | 56.7 |
| 59 | Mistral Large 4Mistral AI | 2026-10-06 | 27 项评测 | ¥0.47 | ¥4.67 | ¥14 | 55.8 |
| 60 | Qwen3.8-27BAlibaba | 2026-08-14 | 34 项评测 | ¥0.6 | ¥3 | ¥12 | 55.7 |
| 61 | Gemini 3.1 Pro PreviewGoogle | 2026-02-19 | 39 项评测 | ¥1.33 | ¥13.33 | ¥79.99 | 55.5 |
| 62 | GLM-5.2Z.ai | 2026-06-16 | 32 项评测 | ¥2 | ¥8 | ¥28 | 54.9 |
| 63 | Qwen3.7 MaxAlibaba | 2026-05-19 | 27 项评测 | ¥2.4 | ¥12 | ¥36 | 54.3 |
| 64 | Claude Sonnet 4.6Anthropic | 2026-02-17 | 30 项评测 | ¥2 | ¥20 | ¥99.99 | 53.5 |
| 65 | Beam 1.0Reflection | 2026-10-16 | 30 项评测 | — | — | — | 52.7 |
| 66 | DeepSeek V4 Pro PreviewDeepSeek | 2026-04-24 | 34 项评测 | ¥0.025 | ¥3 | ¥6 | 52.6 |
| 67 | Kimi K2.6Moonshot AI | 2026-04-20 | 28 项评测 | ¥1.1 | ¥6.5 | ¥27 | 51.2 |
| 68 | MiniMax-M3MiniMax | 2026-06-01 | 44 项评测 | ¥0.42 | ¥2.1 | ¥8.4 | 51.1 |
| 69 | Kimi K2.7 CodeMoonshot AI | 2026-06-12 | 23 项评测 | ¥1.3 | ¥6.5 | ¥27 | 50.8 |
| 70 | Qwen3.6 PlusAlibaba | 2026-04-01 | 23 项评测 | ¥0.2 | ¥2 | ¥12 | 48.5 |
| 71 | GLM-5.1Z.ai | 2026-04-07 | 21 项评测 | ¥1.3 | ¥6 | ¥24 | 48.3 |
| 72 | Qwen3.7 PlusAlibaba | 2026-06-01 | 16 项评测 | ¥0.4 | ¥2 | ¥8 | 48.0 |
| 73 | DeepSeek V4 Flash PreviewDeepSeek | 2026-04-24 | 18 项评测 | ¥0.02 | ¥1 | ¥2 | 47.6 |
| 74 | MiMo-V2.5-ProXiaomi | 2026-04-22 | 19 项评测 | ¥0.025 | ¥3 | ¥6 | 47.5 |
| 75 | GPT-5.4 Mini 2026-03-17OpenAI | 2026-03-17 | 31 项评测 | ¥0.5 | ¥5 | ¥30 | 47.3 |
| 76 | InklingThinking Machines Lab | 2026-07-15 | 44 项评测 | ¥1.13 | ¥6.67 | ¥27 | 47.0 |
| 77 | Gemini 3.5 Flash LiteGoogle | 2026-07-21 | 38 项评测 | ¥0.2 | ¥2 | ¥16.67 | 46.4 |
| 78 | Nemotron 3.5 Super VLNVIDIA | 2026-10-15 | 18 项评测 | — | — | — | 45.8 |
| 79 | Qwen3.6-27BAlibaba | 2026-04-22 | 18 项评测 | — | ¥3 | ¥18 | 44.1 |
| 80 | Inkling-SmallThinking Machines Lab | 2026-07-30 | 34 项评测 | ¥0.4 | ¥2 | ¥8 | 44.1 |
| 81 | Grok 4.3xAI | 2026-04-17 | 28 项评测 | ¥1.33 | ¥8.33 | ¥16.67 | 43.8 |
| 82 | MiMo-V2.5Xiaomi | 2026-04-22 | 15 项评测 | ¥0.02 | ¥1 | ¥2 | 43.8 |
| 83 | Nemotron 3 UltraNVIDIA | 2026-06-04 | 26 项评测 | — | — | — | 43.6 |
| 84 | GPT-5.4 Nano 2026-03-17OpenAI | 2026-03-17 | 26 项评测 | ¥0.13 | ¥1.33 | ¥8.33 | 43.1 |
| 85 | Muse GlimmerMeta | 2026-08-10 | 15 项评测 | — | — | — | 39.6 |
| 86 | MiniMax M 2.7MiniMax | 2026-03-18 | 20 项评测 | ¥0.42 | ¥2.1 | ¥8.4 | 39.0 |
| 87 | Gemma 4 31 BGoogle | 2026-04-02 | 14 项评测 | 免费 | 免费 | 免费 | 37.1 |
| 88 | GPT-5 MiniOpenAI | 2025-08-07 | 17 项评测 | ¥0.17 | ¥1.67 | ¥13.33 | 36.2 |
| 89 | Claude Haiku 4.5 20251001Anthropic | 2025-10-15 | 12 项评测 | ¥0.67 | ¥6.67 | ¥33.33 | 33.6 |
| 90 | Gemini 2.5 ProGoogle | 2025-06-17 | 16 项评测 | ¥0.83 | ¥8.33 | ¥66.66 | 33.2 |
| 91 | Mistral Medium 3.5Mistral AI | 2026-04-29 | 21 项评测 | ¥1 | ¥10 | ¥50 | 31.7 |
| 92 | Gemini 3.1 Flash Lite PreviewGoogle | 2026-03-03 | 23 项评测 | ¥0.17 | ¥1.67 | ¥10 | 30.5 |
| 93 | Granite 4.2 8 BIBM | 2026-08-25 | 13 项评测 | — | — | — | 22.2 |
| 94 | gpt-oss-120bOpenAI | 2025-08-05 | 18 项评测 | — | — | — | 20.8 |
| 95 | Granite 4.2 3 BIBM | 2026-08-25 | 13 项评测 | — | — | — | 11.2 |