1
G
gemini-3.5-flash
gemini-3.5-flashGoogle
100.0
526
1.05M
¥10.8 / ¥64.8输入/输出
对话 · 文本 · 数学排行榜
基于公开偏好评测数据展示 文本 / 数学 榜单，每个榜单独立排序。
数学模型怎么选？
数学排行榜用于回答“数学哪个 AI 模型更强”这类选型问题。页面基于公开偏好评测数据展示模型排名，每个榜单独立排序，不把不同任务的原始分数直接混合。当前榜首是 gemini-3.5-flash，厂商为 google，百分制分数 100.0，样本量 526。 数据来源保留 Arena 原始榜单和 LMArena leaderboard dataset，更新时间以榜单发布时间 2026-05-27 为准。
gemini-3.5-flashclaude-opus-4-6-thinkingclaude-opus-4-6gpt-5.4-highqwen3.7-max-preview
2
A
claude-opus-4-6-thinking
claude-opus-4-6-thinkingAnthropic
99.7
2K
1M
¥36 / ¥180输入/输出
3
Aclaude-opus-4-6
claude-opus-4-6Anthropic
99.4
2.2K
1M
¥36 / ¥180输入/输出
4
O
gpt-5.4-high
gpt-5.4-highOpenai
99.1
1.7K
1.05M
¥18 / ¥108输入/输出
5
A
qwen3.7-max-preview
qwen3.7-max-previewAlibaba
98.9
220
1M
¥18 / ¥54输入/输出
6
Ggemini-3.1-pro-preview
gemini-3.1-pro-previewGoogle
98.6
2.6K
1.05M
¥14.4 / ¥86.4输入/输出
7
Aclaude-opus-4-7-thinking
claude-opus-4-7-thinkingAnthropic
98.3
1.2K
1M
¥36 / ¥180输入/输出
8
Aclaude-opus-4-7
claude-opus-4-7Anthropic
98.0
1.2K
1M
¥36 / ¥180输入/输出
9
MI
mimo-v2.5-pro
mimo-v2.5-proXiaomi
97.7
866
1.05M
¥7.2 / ¥21.6输入/输出
10
B
ernie-5.1
ernie-5.1Baidu
97.4
836
119K
¥5.4 / ¥21.6输入/输出
11
Ogpt-5.5
gpt-5.5Openai
97.1
1K
1.05M
¥36 / ¥216输入/输出
12
Ogpt-5.5-high
gpt-5.5-highOpenai
96.8
1K
1.05M
¥36 / ¥216输入/输出
13
Aqwen3.6-max-preview
qwen3.6-max-previewAlibaba
96.6
327
246K
¥9.5 / ¥56.9输入/输出
14
Z
glm-5.1
glm-5.1Zai
96.3
860
200K
¥0 / ¥0输入/输出
15
Aqwen3.5-max-preview
qwen3.5-max-previewAlibaba
96.0
1.3K
-
-
16
Ggemini-3-pro
gemini-3-proGoogle
95.7
2.7K
1.05M
¥14.4 / ¥86.4输入/输出
17
Ggemini-3-flash
gemini-3-flashGoogle
95.4
2K
1.05M
¥3.6 / ¥21.6输入/输出
18
M
kimi-k2.6
kimi-k2.6Moonshot
95.1
887
262K
¥6.84 / ¥28.8输入/输出
19
Mkimi-k2.5-thinking
kimi-k2.5-thinkingMoonshot
94.8
2.3K
262K
¥4.32 / ¥21.6输入/输出
20
Ggemma-4-26b-a4b
gemma-4-26b-a4bGoogle
94.5
369
262K
¥0.94 / ¥2.88输入/输出
21
D
deepseek-v4-pro-thinking
deepseek-v4-pro-thinkingDeepseek
94.3
886
1M
¥3.13 / ¥6.26输入/输出
22
Ggemma-4-31b
gemma-4-31bGoogle
94.0
398
262K
¥3.24 / ¥7.2输入/输出
23
X
grok-4.20-beta-0309-reasoning
grok-4.20-beta-0309-reasoningXai
93.7
1.7K
2M
¥14.4 / ¥43.2输入/输出
24
Aclaude-opus-4-5-20251101
claude-opus-4-5-20251101Anthropic
93.4
4.1K
200K
¥36 / ¥180输入/输出
25
Aclaude-opus-4-5-20251101-thinking-32k
claude-opus-4-5-20251101-thinking-32kAnthropic
93.1
2.3K
200K
¥108 / ¥540输入/输出
26
Aclaude-sonnet-4-6
claude-sonnet-4-6Anthropic
92.8
1.7K
1M
¥21.6 / ¥108输入/输出
27
M
muse-spark
muse-sparkMeta
92.5
795
-
-
28
Ogpt-5.4
gpt-5.4Openai
92.2
1.8K
1.05M
¥18 / ¥108输入/输出
29
Aqwen3.6-plus
qwen3.6-plusAlibaba
92.0
1.1K
1M
¥3.6 / ¥21.6输入/输出
30
Ggemini-2.5-pro
gemini-2.5-proGoogle
91.7
7.5K
1.05M
¥9 / ¥72输入/输出
31
Aqwen3-max-preview
qwen3-max-previewAlibaba
91.4
1.5K
262K
¥6.2 / ¥24.8输入/输出
32
Ggemini-3-flash (thinking-minimal)
gemini-3-flash (thinking-minimal)Google
91.1
3.2K
1.05M
¥3.6 / ¥21.6输入/输出
33
MImimo-v2-pro
mimo-v2-proXiaomi
90.8
1.5K
1.05M
¥7.2 / ¥21.6输入/输出
34
Aqwen3.5-397b-a17b
qwen3.5-397b-a17bAlibaba
90.5
2K
262K
¥3.1 / ¥18.6输入/输出
35
Aclaude-sonnet-4-5-20250929-thinking-32k
claude-sonnet-4-5-20250929-thinking-32kAnthropic
90.2
4.7K
200K
¥21.6 / ¥108输入/输出
36
Ddeepseek-v4-flash
deepseek-v4-flashDeepseek
89.9
992
1M
¥1.01 / ¥2.02输入/输出
37
Xgrok-4.20-multi-agent-beta-0309
grok-4.20-multi-agent-beta-0309Xai
89.7
1.7K
2M
¥14.4 / ¥43.2输入/输出
38
Ogpt-5.1-high
gpt-5.1-highOpenai
89.4
2.5K
400K
¥9 / ¥72输入/输出
39
Ogpt-5.2-high
gpt-5.2-highOpenai
89.1
2.9K
400K
¥12.6 / ¥101输入/输出
40
Aqwen3-next-80b-a3b-instruct
qwen3-next-80b-a3b-instructAlibaba
88.8
1.2K
131K
¥1.04 / ¥4.13输入/输出
41
Mkimi-k2.5-instant
kimi-k2.5-instantMoonshot
88.5
515
262K
¥4.32 / ¥21.6输入/输出
42
M
longcat-flash-chat
longcat-flash-chatMeituan
88.2
689
128K
¥1.08 / ¥10.8输入/输出
43
A
amazon-nova-experimental-chat-26-02-10
amazon-nova-experimental-chat-26-02-10Amazon
87.9
207
-
-
44
Bernie-5.0-0110
ernie-5.0-0110Baidu
87.6
2.1K
128K
¥7.92 / ¥14.4输入/输出
45
Aqwen3-max-2025-09-23
qwen3-max-2025-09-23Alibaba
87.4
584
258K
¥6.19 / ¥24.7输入/输出
46
MImimo-v2.5
mimo-v2.5Xiaomi
87.1
914
1.05M
¥2.88 / ¥14.4输入/输出
47
B
dola-seed-2.0-pro
dola-seed-2.0-proBytedance
86.8
2.3K
-
-
48
Ogpt-5.2-chat-latest-20260210
gpt-5.2-chat-latest-20260210Openai
86.5
2K
400K
¥12.6 / ¥101输入/输出
49
Ddeepseek-v3.2
deepseek-v3.2Deepseek
86.2
3K
128K
¥2.09 / ¥3.1输入/输出
50
Xgrok-4.20-beta1
grok-4.20-beta1Xai
85.9
1.5K
2M
¥14.4 / ¥43.2输入/输出
51
Zglm-5
glm-5Zai
85.6
1.4K
205K
¥7.2 / ¥23输入/输出
52
Mlongcat-flash-chat-2602-exp
longcat-flash-chat-2602-expMeituan
85.3
1.5K
128K
¥1.08 / ¥10.8输入/输出
53
Zglm-4.6
glm-4.6Zai
85.1
2.1K
205K
¥4.32 / ¥15.8输入/输出
54
Mkimi-k2-thinking-turbo
kimi-k2-thinking-turboMoonshot
84.8
3.7K
262K
¥17.3 / ¥72输入/输出
55
Aqwen3.5-27b
qwen3.5-27bAlibaba
84.5
1.6K
262K
¥2.16 / ¥17.3输入/输出
56
Ddeepseek-v4-pro
deepseek-v4-proDeepseek
84.2
1K
1M
¥3.13 / ¥6.26输入/输出
57
Aamazon-nova-experimental-chat-11-10
amazon-nova-experimental-chat-11-10Amazon
83.9
1.6K
-
-
58
Aqwen3-235b-a22b-instruct-2507
qwen3-235b-a22b-instruct-2507Alibaba
83.6
5.8K
128K
¥2.09 / ¥8.23输入/输出
59
Aclaude-opus-4-1-20250805-thinking-16k
claude-opus-4-1-20250805-thinking-16kAnthropic
83.3
3K
200K
¥108 / ¥540输入/输出
60
Ggemini-3.1-flash-lite-preview
gemini-3.1-flash-lite-previewGoogle
83.0
2.2K
1.05M
¥1.8 / ¥10.8输入/输出
61
Aamazon-nova-experimental-chat-10-20
amazon-nova-experimental-chat-10-20Amazon
82.8
805
-
-
62
Zglm-4.5
glm-4.5Zai
82.5
1.4K
131K
¥4.32 / ¥15.8输入/输出
63
Ogpt-5.5-instant
gpt-5.5-instantOpenai
82.2
1.4K
400K
¥9 / ¥72输入/输出
64
Aqwen3.5-122b-a10b
qwen3.5-122b-a10bAlibaba
81.9
1.7K
262K
¥2.88 / ¥23输入/输出
65
Ddeepseek-v3.2-exp-thinking
deepseek-v3.2-exp-thinkingDeepseek
81.6
481
128K
¥0 / ¥0输入/输出
66
Ddeepseek-v4-flash-thinking
deepseek-v4-flash-thinkingDeepseek
81.3
948
1M
¥1.01 / ¥2.02输入/输出
67
Oo3-2025-04-16
o3-2025-04-16Openai
81.0
3.7K
200K
¥14.4 / ¥57.6输入/输出
68
Xgrok-4-0709
grok-4-0709Xai
80.7
2.3K
256K
¥21.6 / ¥108输入/输出
69
Aqwen3-vl-235b-a22b-instruct
qwen3-vl-235b-a22b-instructAlibaba
80.5
704
128K
¥2.16 / ¥8.64输入/输出
70
Xgrok-4.1-thinking
grok-4.1-thinkingXai
80.2
3.7K
200K
¥14.4 / ¥72输入/输出
71
Zglm-4.7
glm-4.7Zai
79.9
711
205K
¥0 / ¥0输入/输出
72
Ddeepseek-v3.2-exp
deepseek-v3.2-expDeepseek
79.6
775
128K
¥0 / ¥0输入/输出
73
Aclaude-opus-4-1-20250805
claude-opus-4-1-20250805Anthropic
79.3
4.7K
200K
¥108 / ¥540输入/输出
74
TE
hunyuan-hy3-preview
hunyuan-hy3-previewTencent
79.0
378
256K
¥0 / ¥0输入/输出
75
Aamazon-nova-experimental-chat-12-10
amazon-nova-experimental-chat-12-10Amazon
78.7
234
-
-
76
Ddeepseek-v3.1
deepseek-v3.1Deepseek
78.4
992
128K
¥1.44 / ¥5.04输入/输出
77
Aclaude-sonnet-4-5-20250929
claude-sonnet-4-5-20250929Anthropic
78.2
4.7K
200K
¥21.6 / ¥108输入/输出
78
Xgrok-4.1
grok-4.1Xai
77.9
4.1K
200K
¥14.4 / ¥72输入/输出
79
Ogpt-5.2
gpt-5.2Openai
77.6
2.8K
400K
¥12.6 / ¥101输入/输出
80
Ddeepseek-v3.2-thinking
deepseek-v3.2-thinkingDeepseek
77.3
2.5K
128K
¥2.09 / ¥3.1输入/输出
81
Ogpt-5.4-mini-high
gpt-5.4-mini-highOpenai
77.0
1.6K
400K
¥5.4 / ¥32.4输入/输出
82
Ogpt-5.4-nano-high
gpt-5.4-nano-highOpenai
76.7
1.5K
400K
¥1.44 / ¥9输入/输出
83
Xgrok-4-fast-chat
grok-4-fast-chatXai
76.4
399
2M
¥1.44 / ¥3.6输入/输出
84
Ggemini-2.5-flash-preview-09-2025
gemini-2.5-flash-preview-09-2025Google
76.1
1.9K
1M
¥2.16 / ¥18输入/输出
85
MA
mistral-large-3
mistral-large-3Mistral
75.9
2.7K
262K
¥3.6 / ¥10.8输入/输出
86
Aqwen3-vl-235b-a22b-thinking
qwen3-vl-235b-a22b-thinkingAlibaba
75.6
428
131K
¥2.06 / ¥8.26输入/输出
87
Ddeepseek-v3.1-thinking
deepseek-v3.1-thinkingDeepseek
75.3
665
128K
¥1.44 / ¥5.04输入/输出
88
Aqwen3-235b-a22b-thinking-2507
qwen3-235b-a22b-thinking-2507Alibaba
75.0
490
131K
¥2.07 / ¥8.26输入/输出
89
Ogpt-4.5-preview-2025-02-27
gpt-4.5-preview-2025-02-27Openai
74.7
1.4K
8.19K
¥216 / ¥432输入/输出
90
Ggemini-2.5-flash
gemini-2.5-flashGoogle
74.4
7.8K
1.05M
¥2.16 / ¥18输入/输出
91
M
minimax-m2.7
minimax-m2.7Minimax
74.1
1.4K
205K
¥0 / ¥0输入/输出
92
MAmistral-medium-2508
mistral-medium-2508Mistral
73.9
5.7K
262K
¥2.88 / ¥14.4输入/输出
93
Bernie-5.0-preview-1022
ernie-5.0-preview-1022Baidu
73.6
268
128K
¥7.92 / ¥14.4输入/输出
94
Ogpt-5.1
gpt-5.1Openai
73.3
2.9K
400K
¥9 / ¥72输入/输出
95
TEhunyuan-t1-20250711
hunyuan-t1-20250711Tencent
73.0
236
131K
¥0 / ¥0输入/输出
96
Ogpt-5-chat
gpt-5-chatOpenai
72.7
1.8K
400K
¥9 / ¥72输入/输出
97
Ddeepseek-v3.1-terminus-thinking
deepseek-v3.1-terminus-thinkingDeepseek
72.4
200
128K
¥1.8 / ¥5.04输入/输出
98
Aqwen3.5-flash
qwen3.5-flashAlibaba
72.1
1.9K
1M
¥1.24 / ¥12.4输入/输出
99
Xgrok-4-1-fast-reasoning
grok-4-1-fast-reasoningXai
71.8
3.4K
2M
¥1.44 / ¥3.6输入/输出
100
Aqwen3.5-35b-a3b
qwen3.5-35b-a3bAlibaba
71.6
1.7K
262K
¥1.8 / ¥14.4输入/输出
101
Ochatgpt-4o-latest-20250326
chatgpt-4o-latest-20250326Openai
71.3
5.7K
128K
¥18 / ¥72输入/输出
102
Bernie-5.0-preview-1203
ernie-5.0-preview-1203Baidu
71.0
618
128K
¥7.92 / ¥14.4输入/输出
103
ST
step-3.5-flash
step-3.5-flashStepfun
70.7
2.1K
256K
¥0.69 / ¥2.07输入/输出
104
Xgrok-4-fast-reasoning
grok-4-fast-reasoningXai
70.4
1.1K
2M
¥1.44 / ¥3.6输入/输出
105
Ddeepseek-r1-0528
deepseek-r1-0528Deepseek
70.1
869
164K
¥3.6 / ¥15.5输入/输出
106
Aamazon-nova-experimental-chat-26-01-10
amazon-nova-experimental-chat-26-01-10Amazon
69.8
263
-
-
107
Ddeepseek-v3.1-terminus
deepseek-v3.1-terminusDeepseek
69.5
218
128K
¥1.8 / ¥5.04输入/输出
108
Aqwen3-235b-a22b-no-thinking
qwen3-235b-a22b-no-thinkingAlibaba
69.3
2.4K
131K
¥2.07 / ¥8.26输入/输出
109
Xgrok-4.3
grok-4.3Xai
69.0
846
1M
¥9 / ¥18输入/输出
110
Aqwen3-32b
qwen3-32bAlibaba
68.7
316
131K
¥2.07 / ¥8.26输入/输出
111
Ogpt-5-high
gpt-5-highOpenai
68.4
1.9K
400K
¥9 / ¥72输入/输出
112
Zglm-4.5-air
glm-4.5-airZai
68.1
1.5K
131K
¥0 / ¥0输入/输出
113
Mkimi-k2-0905-preview
kimi-k2-0905-previewMoonshot
67.8
759
262K
¥4.32 / ¥18输入/输出
114
MImimo-v2-flash (non-thinking)
mimo-v2-flash (non-thinking)Xiaomi
67.5
2.7K
262K
¥0.72 / ¥2.16输入/输出
115
Oo3-mini-high
o3-mini-highOpenai
67.2
1.9K
200K
¥7.92 / ¥31.7输入/输出
116
Aqwen3-235b-a22b
qwen3-235b-a22bAlibaba
67.0
1.6K
131K
¥2.07 / ¥8.26输入/输出
117
Mminimax-m2.1-preview
minimax-m2.1-previewMinimax
66.7
1K
205K
¥0 / ¥0输入/输出
118
Aqwen3-next-80b-a3b-thinking
qwen3-next-80b-a3b-thinkingAlibaba
66.4
829
131K
¥1.04 / ¥10.3输入/输出
119
Aqwen3-30b-a3b-instruct-2507
qwen3-30b-a3b-instruct-2507Alibaba
66.1
1.4K
262K
¥2.16 / ¥3.6输入/输出
120
NV
nvidia-llama-3.3-nemotron-super-49b-v1.5
nvidia-llama-3.3-nemotron-super-49b-v1.5Nvidia
65.8
194
131K
¥2.88 / ¥2.88输入/输出
121
Ddeepseek-r1
deepseek-r1Deepseek
65.5
1.6K
164K
¥5.04 / ¥18输入/输出
122
Aclaude-opus-4-20250514-thinking-16k
claude-opus-4-20250514-thinking-16kAnthropic
65.2
2.2K
200K
¥108 / ¥540输入/输出
123
Xgrok-3-preview-02-24
grok-3-preview-02-24Xai
64.9
2.7K
1M
¥9 / ¥18输入/输出
124
Aclaude-haiku-4-5-20251001
claude-haiku-4-5-20251001Anthropic
64.7
4.7K
200K
¥7.2 / ¥36输入/输出
125
Oo1-2024-12-17
o1-2024-12-17Openai
64.4
3K
128K
¥108 / ¥432输入/输出
126
Ogpt-oss-120b
gpt-oss-120bOpenai
64.1
1.8K
131K
¥1.08 / ¥4.32输入/输出
127
Oo4-mini-2025-04-16
o4-mini-2025-04-16Openai
63.8
2.9K
200K
¥7.92 / ¥31.7输入/输出
128
Ogpt-5.3-chat-latest
gpt-5.3-chat-latestOpenai
63.5
1.9K
128K
¥12.6 / ¥101输入/输出
129
UNintellect-3
-
63.2
332
131K
¥1.44 / ¥7.92输入/输出
130
Xgrok-3-mini-high
grok-3-mini-highXai
62.9
977
128K
¥0 / ¥0输入/输出
131
Mminimax-m2.5
minimax-m2.5Minimax
62.6
2.2K
205K
¥0 / ¥0输入/输出
132
NVnvidia-nemotron-3-super-120b-a12b
nvidia-nemotron-3-super-120b-a12bNvidia
62.4
511
262K
¥1.44 / ¥5.76输入/输出
133
MImimo-v2-flash (thinking)
mimo-v2-flash (thinking)Xiaomi
62.1
633
262K
¥0.72 / ¥2.16输入/输出
134
Ogpt-5-mini-high
gpt-5-mini-highOpenai
61.8
1.5K
400K
¥1.8 / ¥14.4输入/输出
135
Aclaude-sonnet-4-20250514-thinking-32k
claude-sonnet-4-20250514-thinking-32kAnthropic
61.5
2K
200K
¥21.6 / ¥108输入/输出
136
Ddeepseek-v3-0324
deepseek-v3-0324Deepseek
61.2
3.2K
75K
¥1.44 / ¥5.76输入/输出
137
NVnvidia-nemotron-3-nano-30b-a3b-bf16
nvidia-nemotron-3-nano-30b-a3b-bf16Nvidia
60.9
987
131K
¥0 / ¥0输入/输出
138
Ggemini-2.5-flash-lite-preview-09-2025-no-thinking
gemini-2.5-flash-lite-preview-09-2025-no-thinkingGoogle
60.6
2.9K
1.05M
¥0.72 / ¥2.88输入/输出
139
Oo3-mini
o3-miniOpenai
60.3
4.7K
200K
¥7.92 / ¥31.7输入/输出
140
Aclaude-opus-4-20250514
claude-opus-4-20250514Anthropic
60.1
2.8K
200K
¥108 / ¥540输入/输出
141
Oo1-preview
o1-previewOpenai
59.8
4.6K
128K
¥108 / ¥432输入/输出
142
UNtrinity-large-thinking
-
59.5
1.4K
262K
¥1.8 / ¥6.48输入/输出
143
AG
ling-flash-2.0
ling-flash-2.0Ant Group
59.2
461
131K
¥1.01 / ¥4.1输入/输出
144
Xgrok-3-mini-beta
grok-3-mini-betaXai
58.9
1.5K
1M
¥9 / ¥18输入/输出
145
Aqwen2.5-max
qwen2.5-maxAlibaba
58.6
3.3K
32K
¥11.5 / ¥46输入/输出
146
Ogpt-4.1-2025-04-14
gpt-4.1-2025-04-14Openai
58.3
3.2K
1.05M
¥14.4 / ¥57.6输入/输出
147
Mkimi-k2-0711-preview
kimi-k2-0711-previewMoonshot
58.0
1.7K
131K
¥4.32 / ¥18输入/输出
148
STstep-3
step-3Stepfun
57.8
353
65.5K
¥1.8 / ¥4.68输入/输出
149
Aqwen3-coder-480b-a35b-instruct
qwen3-coder-480b-a35b-instructAlibaba
57.5
1.6K
262K
¥6.2 / ¥24.8输入/输出
150
Ggemini-2.5-flash-lite-preview-06-17-thinking
gemini-2.5-flash-lite-preview-06-17-thinkingGoogle
57.2
2.1K
65.5K
¥0.72 / ¥2.88输入/输出
151
Mminimax-m1
minimax-m1Minimax
56.9
1.8K
1M
¥0.95 / ¥9.03输入/输出
152
Anova-2-lite
nova-2-liteAmazon
56.6
825
128K
¥2.38 / ¥19.8输入/输出
153
NVllama-3.1-nemotron-ultra-253b-v1
llama-3.1-nemotron-ultra-253b-v1Nvidia
56.3
209
128K
¥4.32 / ¥13输入/输出
154
Aqwq-32b
qwq-32bAlibaba
56.0
1.7K
131K
¥2.07 / ¥6.2输入/输出
155
TEhunyuan-turbos-20250416
hunyuan-turbos-20250416Tencent
55.7
845
131K
¥0 / ¥0输入/输出
156
Zglm-4.7-flash
glm-4.7-flashZai
55.5
718
200K
¥0 / ¥0输入/输出
157
Oo1-mini
o1-miniOpenai
55.2
7.5K
128K
¥7.92 / ¥31.7输入/输出
158
Aclaude-sonnet-4-20250514
claude-sonnet-4-20250514Anthropic
54.9
2.5K
200K
¥21.6 / ¥108输入/输出
159
Aqwen3-30b-a3b
qwen3-30b-a3bAlibaba
54.6
1.7K
128K
¥0.79 / ¥7.78输入/输出
160
Mminimax-m2
minimax-m2Minimax
54.3
318
197K
¥0 / ¥0输入/输出
161
MAmistral-medium-2505
mistral-medium-2505Mistral
54.0
2.2K
262K
¥2.88 / ¥14.4输入/输出
162
Ggemini-2.0-flash-001
gemini-2.0-flash-001Google
53.7
4.1K
1.05M
¥1.08 / ¥4.32输入/输出
163
Zglm-4.5v
glm-4.5vZai
53.4
276
64K
¥4.32 / ¥13输入/输出
164
AGring-flash-2.0
ring-flash-2.0Ant Group
53.2
453
131K
¥1.01 / ¥4.1输入/输出
165
Ogpt-4.1-mini-2025-04-14
gpt-4.1-mini-2025-04-14Openai
52.9
2.7K
1.05M
¥2.88 / ¥11.5输入/输出
166
MAmistral-small-2506
mistral-small-2506Mistral
52.6
1K
262K
¥2.88 / ¥14.4输入/输出
167
Aclaude-3-7-sonnet-20250219-thinking-32k
claude-3-7-sonnet-20250219-thinking-32kAnthropic
52.3
2.8K
-
-
168
UNtrinity-large-preview
-
52.0
1.8K
262K
¥1.8 / ¥6.48输入/输出
169
Aqwen-plus-0125
qwen-plus-0125Alibaba
51.7
732
1M
¥0.83 / ¥2.07输入/输出
170
Aclaude-3-7-sonnet-20250219
claude-3-7-sonnet-20250219Anthropic
51.4
3.4K
200K
¥21.6 / ¥108输入/输出
171
STstep-1o-turbo-202506
step-1o-turbo-202506Stepfun
51.1
564
-
-
172
Ogpt-oss-20b
gpt-oss-20bOpenai
50.9
680
131K
¥0.32 / ¥1.3输入/输出
173
Ogpt-5-nano-high
gpt-5-nano-highOpenai
50.6
494
400K
¥0.36 / ¥2.88输入/输出
174
AI
olmo-3-32b-think
olmo-3-32b-thinkAllenai
50.3
314
128K
¥2.16 / ¥3.24输入/输出
175
Ggemini-1.5-pro-002
gemini-1.5-pro-002Google
50.0
7.6K
-
-
176
Ggemma-3-27b-it
gemma-3-27b-itGoogle
49.7
3.6K
128K
¥2.15 / ¥2.15输入/输出
177
AIolmo-3.1-32b-instruct
olmo-3.1-32b-instructAllenai
49.4
696
200K
¥14.4 / ¥57.6输入/输出
178
Ddeepseek-v3
deepseek-v3Deepseek
49.1
2.7K
128K
¥0 / ¥0输入/输出
179
Ggemini-2.0-flash-lite-preview-02-05
gemini-2.0-flash-lite-preview-02-05Google
48.9
2.8K
1.05M
¥0.54 / ¥2.16输入/输出
180
IB
granite-4.1-8b
granite-4.1-8bIbm
48.6
218
131K
¥0.36 / ¥0.72输入/输出
181
Ggemma-3-12b-it
gemma-3-12b-itGoogle
48.3
389
128K
¥1.96 / ¥1.96输入/输出
182
Aclaude-3-5-sonnet-20241022
claude-3-5-sonnet-20241022Anthropic
48.0
10K
200K
¥21.6 / ¥108输入/输出
183
STstep-2-16k-exp-202412
step-2-16k-exp-202412Stepfun
47.7
642
16.4K
¥37.5 / ¥118输入/输出
184
Aclaude-3-5-sonnet-20240620
claude-3-5-sonnet-20240620Anthropic
47.4
11.4K
200K
¥21.6 / ¥108输入/输出
185
UNathene-v2-chat
-
47.1
3.4K
-
-
186
Mllama-4-maverick-17b-128e-instruct
llama-4-maverick-17b-128e-instructMeta
46.8
2.8K
1M
¥1.8 / ¥6.26输入/输出
187
UNyi-lightning
-
46.6
3.9K
12K
¥1.44 / ¥1.44输入/输出
188
CO
command-a-03-2025
command-a-03-2025Cohere
46.3
4K
256K
¥18 / ¥72输入/输出
189
AIolmo-3.1-32b-think
olmo-3.1-32b-thinkAllenai
46.0
473
200K
¥14.4 / ¥57.6输入/输出
190
Aqwen2.5-plus-1127
qwen2.5-plus-1127Alibaba
45.7
1.4K
-
-
191
TEhunyuan-turbos-20250226
hunyuan-turbos-20250226Tencent
45.4
238
131K
¥0 / ¥0输入/输出
192
Ddeepseek-v2.5-1210
deepseek-v2.5-1210Deepseek
45.1
1K
1M
¥1.01 / ¥2.02输入/输出
193
Zglm-4-plus-0111
glm-4-plus-0111Zai
44.8
721
128K
¥72 / ¥72输入/输出
194
Mllama-4-scout-17b-16e-instruct
llama-4-scout-17b-16e-instructMeta
44.5
1.9K
128K
¥1.44 / ¥5.62输入/输出
195
Ogpt-4o-2024-08-06
gpt-4o-2024-08-06Openai
44.3
6.8K
128K
¥18 / ¥72输入/输出
196
Ogpt-4o-2024-05-13
gpt-4o-2024-05-13Openai
44.0
15.1K
128K
¥36 / ¥108输入/输出
197
Xgrok-2-2024-08-13
grok-2-2024-08-13Xai
43.7
9K
1M
¥9 / ¥18输入/输出
198
Aqwen2.5-72b-instruct
qwen2.5-72b-instructAlibaba
43.4
5.4K
131K
¥4.13 / ¥12.4输入/输出
199
Mllama-3.1-405b-instruct-fp8
llama-3.1-405b-instruct-fp8Meta
43.1
8.5K
128K
¥0 / ¥0输入/输出
200
TEhunyuan-large-2025-02-10
hunyuan-large-2025-02-10Tencent
42.8
497
-
-
201
Mllama-3.1-405b-instruct-bf16
llama-3.1-405b-instruct-bf16Meta
42.5
5.2K
128K
¥0 / ¥0输入/输出
202
Aqwen-max-0919
qwen-max-0919Alibaba
42.2
2.2K
131K
¥2.48 / ¥9.91输入/输出
203
Zglm-4-plus
glm-4-plusZai
42.0
3.6K
128K
¥54 / ¥54输入/输出
204
Ogpt-4.1-nano-2025-04-14
gpt-4.1-nano-2025-04-14Openai
41.7
582
1.05M
¥14.4 / ¥57.6输入/输出
205
TEhunyuan-standard-2025-02-10
hunyuan-standard-2025-02-10Tencent
41.4
499
-
-
206
TEhunyuan-turbo-0110
hunyuan-turbo-0110Tencent
41.1
243
-
-
207
Aclaude-3-opus-20240229
claude-3-opus-20240229Anthropic
40.8
25.8K
200K
¥108 / ¥540输入/输出
208
Ggemini-advanced-0514
gemini-advanced-0514Google
40.5
6.4K
-
-
209
Ogpt-4-turbo-2024-04-09
gpt-4-turbo-2024-04-09Openai
40.2
13.2K
128K
¥72 / ¥216输入/输出
210
NVllama-3.1-nemotron-70b-instruct
llama-3.1-nemotron-70b-instructNvidia
39.9
1K
128K
¥0 / ¥0输入/输出
211
Ddeepseek-v2.5
deepseek-v2.5Deepseek
39.7
3.6K
1M
¥1.01 / ¥2.02输入/输出
212
Ggemini-1.5-pro-001
gemini-1.5-pro-001Google
39.4
10.5K
-
-
213
Ogpt-4-1106-preview
gpt-4-1106-previewOpenai
39.1
13.3K
8.19K
¥216 / ¥432输入/输出
214
Ggemini-1.5-flash-002
gemini-1.5-flash-002Google
38.8
4.8K
2M
¥0.54 / ¥2.2输入/输出
215
TEhunyuan-large-vision
hunyuan-large-visionTencent
38.5
351
-
-
216
Ogpt-4-0125-preview
gpt-4-0125-previewOpenai
38.2
12.4K
8.19K
¥216 / ¥432输入/输出
217
Ogpt-4o-mini-2024-07-18
gpt-4o-mini-2024-07-18Openai
37.9
9.3K
128K
¥1.08 / ¥4.32输入/输出
218
Mllama-3.3-70b-instruct
llama-3.3-70b-instructMeta
37.6
5.8K
128K
¥0 / ¥0输入/输出
219
Xgrok-2-mini-2024-08-13
grok-2-mini-2024-08-13Xai
37.4
7.3K
1M
¥9 / ¥18输入/输出
220
MAmistral-large-2407
mistral-large-2407Mistral
37.1
6.7K
131K
¥14.4 / ¥43.2输入/输出
221
MAmistral-small-3.1-24b-instruct-2503
mistral-small-3.1-24b-instruct-2503Mistral
36.8
2.1K
262K
¥2.88 / ¥14.4输入/输出
222
MAmistral-large-2411
mistral-large-2411Mistral
36.5
3.6K
128K
¥14.4 / ¥43.2输入/输出
223
Mllama-3.1-70b-instruct
llama-3.1-70b-instructMeta
36.2
7.7K
131K
¥2.88 / ¥2.88输入/输出
224
Aamazon-nova-pro-v1.0
amazon-nova-pro-v1.0Amazon
35.9
3K
300K
¥5.76 / ¥23输入/输出
225
IBibm-granite-h-small
ibm-granite-h-smallIbm
35.6
358
-
-
226
Ggemma-3n-e4b-it
gemma-3n-e4b-itGoogle
35.3
1.6K
128K
¥0 / ¥0输入/输出
227
Aqwen2.5-coder-32b-instruct
qwen2.5-coder-32b-instructAlibaba
35.1
725
131K
¥2.07 / ¥6.2输入/输出
228
MAmagistral-medium-2506
magistral-medium-2506Mistral
34.8
553
128K
¥14.4 / ¥36输入/输出
229
M
phi-4
phi-4Microsoft
34.5
2.8K
128K
¥0.9 / ¥3.6输入/输出
230
Aclaude-3-5-haiku-20241022
claude-3-5-haiku-20241022Anthropic
34.2
6.4K
200K
¥5.76 / ¥28.8输入/输出
231
AIllama-3.1-tulu-3-70b
llama-3.1-tulu-3-70bAllenai
33.9
397
-
-
232
Ddeepseek-coder-v2
deepseek-coder-v2Deepseek
33.6
1.9K
1M
¥1.01 / ¥2.02输入/输出
233
MAmistral-small-24b-instruct-2501
mistral-small-24b-instruct-2501Mistral
33.3
1.7K
262K
¥2.88 / ¥14.4输入/输出
234
Ggemma-3-4b-it
gemma-3-4b-itGoogle
33.0
423
128K
¥1.44 / ¥1.44输入/输出
235
Aqwen2-72b-instruct
qwen2-72b-instructAlibaba
32.8
4.8K
131K
¥4.13 / ¥12.4输入/输出
236
TEhunyuan-standard-256k
hunyuan-standard-256kTencent
32.5
361
-
-
237
UNathene-70b-0725
-
32.2
2.9K
-
-
238
Ogpt-4-0314
gpt-4-0314Openai
31.9
7.1K
8.19K
¥216 / ¥432输入/输出
239
NVllama-3.1-nemotron-51b-instruct
llama-3.1-nemotron-51b-instructNvidia
31.6
507
128K
¥0 / ¥0输入/输出
240
Ggemini-1.5-flash-001
gemini-1.5-flash-001Google
31.3
8.4K
2M
¥0.54 / ¥2.2输入/输出
241
Aamazon-nova-lite-v1.0
amazon-nova-lite-v1.0Amazon
31.0
2.5K
300K
¥0.43 / ¥1.73输入/输出
242
UNreka-core-20240904
-
30.7
1.2K
-
-
243
UNjamba-1.5-large
-
30.5
1.1K
256K
¥0 / ¥0输入/输出
244
Zglm-4-0520
glm-4-0520Zai
30.2
1.2K
128K
¥108 / ¥108输入/输出
245
Mllama-3-70b-instruct
llama-3-70b-instructMeta
29.9
20.9K
8.19K
¥3.67 / ¥5.33输入/输出
246
Ogpt-4-0613
gpt-4-0613Openai
29.6
11.2K
8.19K
¥216 / ¥432输入/输出
247
NVnemotron-4-340b-instruct
nemotron-4-340b-instructNvidia
29.3
2.4K
-
-
248
Aqwq-32b-preview
qwq-32b-previewAlibaba
29.0
480
131K
¥2.07 / ¥6.2输入/输出
249
Aclaude-3-sonnet-20240229
claude-3-sonnet-20240229Anthropic
28.7
13.8K
200K
¥21.6 / ¥108输入/输出
250
Ggemma-2-27b-it
gemma-2-27b-itGoogle
28.4
10.2K
8.19K
¥0.58 / ¥0.58输入/输出
251
AIolmo-2-0325-32b-instruct
olmo-2-0325-32b-instructAllenai
28.2
375
-
-
252
Ggemini-1.5-flash-8b-001
gemini-1.5-flash-8b-001Google
27.9
5K
2M
¥0.54 / ¥2.2输入/输出
253
Aamazon-nova-micro-v1.0
amazon-nova-micro-v1.0Amazon
27.6
2.5K
128K
¥0.25 / ¥1.01输入/输出
254
MAmistral-large-2402
mistral-large-2402Mistral
27.3
8K
262K
¥2.88 / ¥14.4输入/输出
255
COc4ai-aya-expanse-32b
c4ai-aya-expanse-32bCohere
27.0
3.9K
-
-
256
UNreka-flash-20240904
-
26.7
1.3K
65.5K
¥0.72 / ¥1.44输入/输出
257
AIllama-3.1-tulu-3-8b
llama-3.1-tulu-3-8bAllenai
26.4
363
-
-
258
MAministral-8b-2410
ministral-8b-2410Mistral
26.1
683
128K
¥0.72 / ¥0.72输入/输出
259
Aclaude-3-haiku-20240307
claude-3-haiku-20240307Anthropic
25.9
15K
200K
¥1.8 / ¥9输入/输出
260
COcommand-r-plus-08-2024
command-r-plus-08-2024Cohere
25.6
1.5K
128K
¥18 / ¥72输入/输出
261
Aqwen1.5-110b-chat
qwen1.5-110b-chatAlibaba
25.3
3.2K
-
-
262
MAmixtral-8x22b-instruct-v0.1
mixtral-8x22b-instruct-v0.1Mistral
25.0
6.8K
64K
¥14.4 / ¥43.2输入/输出
263
Ggemma-2-9b-it
gemma-2-9b-itGoogle
24.7
7.1K
8.19K
¥1.44 / ¥1.44输入/输出
264
UNyi-1.5-34b-chat
-
24.4
3K
-
-
265
MAmistral-medium
mistral-mediumMistral
24.1
4.4K
262K
¥2.88 / ¥14.4输入/输出
266
UNinternlm2_5-20b-chat
-
23.9
1.4K
-
-
267
Mllama-3.1-8b-instruct
llama-3.1-8b-instructMeta
23.6
7.1K
131K
¥0.79 / ¥0.79输入/输出
268
Mphi-3-medium-4k-instruct
phi-3-medium-4k-instructMicrosoft
23.3
3.2K
4.1K
¥1.22 / ¥4.9输入/输出
269
UNgemma-2-9b-it-simpo
-
23.0
1.3K
8.19K
¥1.44 / ¥1.44输入/输出
270
COc4ai-aya-expanse-8b
c4ai-aya-expanse-8bCohere
22.7
1.3K
-
-
271
UNreka-flash-21b-20240226-online
-
22.4
2K
-
-
272
COcommand-r-plus
command-r-plusCohere
22.1
9.8K
128K
¥18 / ¥72输入/输出
273
Aqwen1.5-72b-chat
qwen1.5-72b-chatAlibaba
21.8
5.3K
-
-
274
UNjamba-1.5-mini
-
21.6
1.1K
256K
¥0 / ¥0输入/输出
275
IBgranite-3.1-2b-instruct
granite-3.1-2b-instructIbm
21.3
391
-
-
276
UNreka-flash-21b-20240226
-
21.0
3.4K
-
-
277
Aqwen1.5-32b-chat
qwen1.5-32b-chatAlibaba
20.7
2.6K
-
-
278
COcommand-r-08-2024
command-r-08-2024Cohere
20.4
1.6K
128K
¥18 / ¥72输入/输出
279
Mphi-3-mini-4k-instruct-june-2024
phi-3-mini-4k-instruct-june-2024Microsoft
20.1
1.6K
4.1K
¥0.94 / ¥3.74输入/输出
280
IBgranite-3.1-8b-instruct
granite-3.1-8b-instructIbm
19.8
382
-
-
281
Mllama-3-8b-instruct
llama-3-8b-instructMeta
19.5
14.3K
8.19K
¥0.29 / ¥0.29输入/输出
282
Mphi-3-small-8k-instruct
phi-3-small-8k-instructMicrosoft
19.3
2.1K
8.19K
¥1.08 / ¥4.32输入/输出
283
UNzephyr-orpo-141b-A35b-v0.1
-
19.0
589
200K
¥108 / ¥432输入/输出
284
MAmixtral-8x7b-instruct-v0.1
mixtral-8x7b-instruct-v0.1Mistral
18.7
9.7K
32K
¥5.04 / ¥5.04输入/输出
285
UNdbrx-instruct-preview
-
18.4
4K
-
-
286
IBgranite-3.0-8b-instruct
granite-3.0-8b-instructIbm
18.1
873
-
-
287
Ogpt-3.5-turbo-0125
gpt-3.5-turbo-0125Openai
17.8
8.6K
16.4K
¥3.6 / ¥10.8输入/输出
288
Ogpt-3.5-turbo-1106
gpt-3.5-turbo-1106Openai
17.5
2.1K
16.4K
¥7.2 / ¥14.4输入/输出
289
Ggemma-2-2b-it
gemma-2-2b-itGoogle
17.2
6.6K
128K
¥0 / ¥0输入/输出
290
Ggemini-pro-dev-api
gemini-pro-dev-apiGoogle
17.0
2.3K
1.05M
¥14.4 / ¥86.4输入/输出
291
Ggemini-pro
gemini-proGoogle
16.7
993
1.05M
¥14.4 / ¥86.4输入/输出
292
Mllama-3.2-3b-instruct
llama-3.2-3b-instructMeta
16.4
1.1K
131K
¥0.22 / ¥0.35输入/输出
293
Aqwen1.5-14b-chat
qwen1.5-14b-chatAlibaba
16.1
2.2K
-
-
294
UNstarling-lm-7b-beta
-
15.8
2K
200K
¥5.4 / ¥18.7输入/输出
295
COcommand-r
command-rCohere
15.5
6.7K
128K
¥18 / ¥72输入/输出
296
IBgranite-3.0-2b-instruct
granite-3.0-2b-instructIbm
15.2
908
-
-
297
Mwizardlm-70b
wizardlm-70bMicrosoft
14.9
903
-
-
298
UNyi-34b-chat
-
14.7
2K
-
-
299
Mphi-3-mini-4k-instruct
phi-3-mini-4k-instructMicrosoft
14.4
2.6K
4.1K
¥0.94 / ¥3.74输入/输出
300
UNsnowflake-arctic-instruct
-
14.1
4.8K
-
-
301
Ddeepseek-llm-67b-chat
deepseek-llm-67b-chatDeepseek
13.8
576
1M
¥1.01 / ¥2.02输入/输出
302
UNtulu-2-dpo-70b
-
13.5
888
-
-
303
Ggemma-1.1-7b-it
gemma-1.1-7b-itGoogle
13.2
3K
-
-
304
UNopenchat-3.5-0106
-
12.9
1.7K
-
-
305
UNsmollm2-1.7b-instruct
-
12.6
271
-
-
306
UNopenhermes-2.5-mistral-7b
-
12.4
697
1M
¥36 / ¥180输入/输出
307
Mllama-2-70b-chat
llama-2-70b-chatMeta
12.1
4.7K
-
-
308
Mphi-3-mini-128k-instruct
phi-3-mini-128k-instructMicrosoft
11.8
2.8K
128K
¥0.94 / ¥3.74输入/输出
309
Mllama-3.2-1b-instruct
llama-3.2-1b-instructMeta
11.5
1.2K
16.4K
¥0.07 / ¥0.08输入/输出
310
MAmistral-7b-instruct-v0.2
mistral-7b-instruct-v0.2Mistral
11.2
2.6K
262K
¥2.88 / ¥14.4输入/输出
311
UNstarling-lm-7b-alpha
-
10.9
1.3K
200K
¥5.4 / ¥18.7输入/输出
312
Aqwen1.5-7b-chat
qwen1.5-7b-chatAlibaba
10.6
690
-
-
313
UNdolphin-2.2.1-mistral-7b
-
10.3
219
262K
¥2.88 / ¥14.4输入/输出
314
NVllama2-70b-steerlm-chat
llama2-70b-steerlm-chatNvidia
10.1
440
-
-
315
UNopenchat-3.5
-
9.8
945
-
-
316
UNvicuna-33b
-
9.5
2.7K
-
-
317
Aqwen-14b-chat
qwen-14b-chatAlibaba
9.2
534
32.8K
¥1.04 / ¥3.1输入/输出
318
Ggemma-7b-it
gemma-7b-itGoogle
8.9
1.1K
-
-
319
Mllama-2-13b-chat
llama-2-13b-chatMeta
8.6
2.2K
-
-
320
UNsolar-10.7b-instruct-v1.0
-
8.3
604
128K
¥0 / ¥0输入/输出
321
UNnous-hermes-2-mixtral-8x7b-dpo
-
8.0
628
1M
¥36 / ¥180输入/输出
322
Mcodellama-34b-instruct
codellama-34b-instructMeta
7.8
770
-
-
323
Gpalm-2
palm-2Google
7.5
901
-
-
324
Ggemma-1.1-2b-it
gemma-1.1-2b-itGoogle
7.2
1.4K
-
-
325
UNmpt-30b-chat
-
6.9
242
-
-
326
Mllama-2-7b-chat
llama-2-7b-chatMeta
6.6
1.7K
128K
¥4.03 / ¥48输入/输出
327
UNzephyr-7b-beta
-
6.3
1.3K
-
-
328
UNstripedhyena-nous-7b
-
6.0
676
-
-
329
UNguanaco-33b
-
5.7
280
200K
¥14.4 / ¥57.6输入/输出
330
UNvicuna-13b
-
5.5
2.1K
-
-
331
MAmistral-7b-instruct
mistral-7b-instructMistral
5.2
974
262K
¥2.88 / ¥14.4输入/输出
332
Aqwen1.5-4b-chat
qwen1.5-4b-chatAlibaba
4.9
988
-
-
333
AIolmo-7b-instruct
olmo-7b-instructAllenai
4.6
848
-
-
334
Mwizardlm-13b
wizardlm-13bMicrosoft
4.3
669
-
-
335
Ggemma-2b-it
gemma-2b-itGoogle
4.0
597
-
-
336
UNvicuna-7b
-
3.7
658
-
-
337
UNchatglm3-6b
-
3.4
576
200K
¥5.4 / ¥18.7输入/输出
338
UNgpt4all-13b-snoozy
-
3.2
211
1M
¥36 / ¥216输入/输出
339
UNkoala-13b
-
2.9
751
-
-
340
UNchatglm-6b
-
2.6
525
200K
¥5.4 / ¥18.7输入/输出
341
UNRWKV-4-Raven-14B
-
2.3
544
-
-
342
UNmpt-7b-chat
-
2.0
471
-
-
343
UNchatglm2-6b
-
1.7
227
200K
¥5.4 / ¥18.7输入/输出
344
UNalpaca-13b
-
1.4
652
-
-
345
UNoasst-pythia-12b
-
1.1
687
-
-
346
UNdolly-v2-12b
-
0.9
370
-
-
347
UNfastchat-t5-3b
-
0.6
462
-
-
348
UNstablelm-tuned-alpha-7b
-
0.3
353
-
-
349
Mllama-13b
llama-13bMeta
0.0
252
-
-
gemini-3.5-flash 为什么排第一
gemini-3.5-flash 当前排名第 1，厂商为 google，百分制分数 100.0，样本量 526。如果你的任务与本榜单口径一致，可以优先把它作为候选模型，再结合价格、上下文长度和可用性做最终选择。
不要只看第一名
先选择最接近真实任务的排行榜，再比较 Top 模型的分数和样本量。分数越高说明同一榜单内偏好表现越强；样本量越高通常结果更稳定。最后再结合价格、上下文长度、开源闭源和厂商可用性筛选。
继续比较相近能力
数学常见问题
数学排行榜看什么指标？
主要看排名、百分制分数、样本量和来源。分数用于快速比较同一榜单内模型表现，样本量用于判断结果稳定性。
为什么不同榜单不能直接混合成总分？
不同榜单的任务、样本和评测口径不同，模力榜默认只在同一榜单内排序，避免把写作、代码、图像等能力强行合并。
数学模型应该怎么选？
优先看与你任务最接近的榜单，再结合价格、上下文长度、开源闭源和厂商可用性。排名靠前不代表适合所有预算和部署方式。
榜单多久更新？
页面展示的是最新成功采集的公开榜单数据。当前优先使用 LMArena leaderboard dataset，并在页面来源中保留原始链接。