1
A
claude-opus-4-6
claude-opus-4-6Anthropic
100.0
1.9K
1M
¥36 / ¥180输入/输出
对话 · 文本 · 数学行业任务排行榜
基于公开偏好评测数据展示 文本 / 数学行业任务 榜单，每个榜单独立排序。
数学行业任务模型怎么选？
数学行业任务排行榜用于回答“数学行业任务哪个 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 claude-opus-4-6，厂商为 anthropic，百分制分数 100.0，样本量 1.9K。 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。
claude-opus-4-6claude-opus-4-6-thinkinggemini-3.5-flashmimo-v2.5-progpt-5.4-high
2
Aclaude-opus-4-6-thinking
claude-opus-4-6-thinkingAnthropic
99.7
1.7K
1M
¥36 / ¥180输入/输出
3
G
gemini-3.5-flash
gemini-3.5-flashGoogle
99.4
561
1.05M
¥10.8 / ¥64.8输入/输出
4
MI
mimo-v2.5-pro
mimo-v2.5-proXiaomi
99.1
839
1.05M
¥7.2 / ¥21.6输入/输出
5
O
gpt-5.4-high
gpt-5.4-highOpenai
98.8
1.6K
1.05M
¥18 / ¥108输入/输出
6
Aclaude-opus-4-7-thinking
claude-opus-4-7-thinkingAnthropic
98.5
1.1K
1M
¥36 / ¥180输入/输出
7
Aclaude-opus-4-7
claude-opus-4-7Anthropic
98.2
1.2K
1M
¥36 / ¥180输入/输出
8
Ogpt-5.5
gpt-5.5Openai
97.9
982
1.05M
¥36 / ¥216输入/输出
9
A
qwen3.5-max-preview
qwen3.5-max-previewAlibaba
97.7
1.1K
-
-
10
Ogpt-5.5-high
gpt-5.5-highOpenai
97.4
987
1.05M
¥36 / ¥216输入/输出
11
B
ernie-5.1
ernie-5.1Baidu
97.1
759
119K
¥5.4 / ¥21.6输入/输出
12
Aqwen3.6-max-preview
qwen3.6-max-previewAlibaba
96.8
308
246K
¥9.5 / ¥56.9输入/输出
13
M
kimi-k2.6
kimi-k2.6Moonshot
96.5
838
262K
¥6.84 / ¥28.8输入/输出
14
Ggemini-3.1-pro-preview
gemini-3.1-pro-previewGoogle
96.2
2.3K
1.05M
¥14.4 / ¥86.4输入/输出
15
Aclaude-sonnet-4-6
claude-sonnet-4-6Anthropic
95.9
1.5K
1M
¥21.6 / ¥108输入/输出
16
Ggemini-3-pro
gemini-3-proGoogle
95.6
1.9K
1.05M
¥14.4 / ¥86.4输入/输出
17
Mkimi-k2.5-thinking
kimi-k2.5-thinkingMoonshot
95.3
1.8K
262K
¥4.32 / ¥21.6输入/输出
18
Z
glm-5.1
glm-5.1Zai
95.0
802
200K
¥0 / ¥0输入/输出
19
Aqwen3.7-max-preview
qwen3.7-max-previewAlibaba
94.7
227
1M
¥18 / ¥54输入/输出
20
Aclaude-opus-4-5-20251101
claude-opus-4-5-20251101Anthropic
94.4
3.1K
200K
¥36 / ¥180输入/输出
21
Ggemma-4-26b-a4b
gemma-4-26b-a4bGoogle
94.1
296
262K
¥0.94 / ¥2.88输入/输出
22
Ggemini-3-flash
gemini-3-flashGoogle
93.8
1.4K
1.05M
¥3.6 / ¥21.6输入/输出
23
MImimo-v2-pro
mimo-v2-proXiaomi
93.5
1.3K
1.05M
¥7.2 / ¥21.6输入/输出
24
Ggemma-4-31b
gemma-4-31bGoogle
93.3
322
262K
¥3.24 / ¥7.2输入/输出
25
Aclaude-sonnet-4-5-20250929-thinking-32k
claude-sonnet-4-5-20250929-thinking-32kAnthropic
93.0
3.6K
200K
¥21.6 / ¥108输入/输出
26
Aqwen3.5-397b-a17b
qwen3.5-397b-a17bAlibaba
92.7
1.7K
262K
¥3.1 / ¥18.6输入/输出
27
Ogpt-5.4
gpt-5.4Openai
92.4
1.6K
1.05M
¥18 / ¥108输入/输出
28
D
deepseek-v4-pro-thinking
deepseek-v4-pro-thinkingDeepseek
92.1
836
1M
¥3.13 / ¥6.26输入/输出
29
M
muse-spark
muse-sparkMeta
91.8
717
-
-
30
Aclaude-opus-4-5-20251101-thinking-32k
claude-opus-4-5-20251101-thinking-32kAnthropic
91.5
1.6K
200K
¥108 / ¥540输入/输出
31
X
grok-4.20-beta-0309-reasoning
grok-4.20-beta-0309-reasoningXai
91.2
1.6K
2M
¥14.4 / ¥43.2输入/输出
32
Ddeepseek-v4-pro
deepseek-v4-proDeepseek
90.9
973
1M
¥3.13 / ¥6.26输入/输出
33
Zglm-5
glm-5Zai
90.6
1.2K
205K
¥7.2 / ¥23输入/输出
34
MImimo-v2.5
mimo-v2.5Xiaomi
90.3
914
1.05M
¥2.88 / ¥14.4输入/输出
35
Ggemini-2.5-pro
gemini-2.5-proGoogle
90.0
6.5K
1.05M
¥9 / ¥72输入/输出
36
Aqwen3.6-plus
qwen3.6-plusAlibaba
89.7
1K
1M
¥3.6 / ¥21.6输入/输出
37
Xgrok-4.20-multi-agent-beta-0309
grok-4.20-multi-agent-beta-0309Xai
89.4
1.6K
2M
¥14.4 / ¥43.2输入/输出
38
Zglm-4.6
glm-4.6Zai
89.1
1.7K
205K
¥4.32 / ¥15.8输入/输出
39
Aqwen3-max-preview
qwen3-max-previewAlibaba
88.9
1.3K
262K
¥6.2 / ¥24.8输入/输出
40
Ogpt-5.1-high
gpt-5.1-highOpenai
88.6
1.7K
400K
¥9 / ¥72输入/输出
41
Aclaude-sonnet-4-5-20250929
claude-sonnet-4-5-20250929Anthropic
88.3
3.7K
200K
¥21.6 / ¥108输入/输出
42
Ddeepseek-v3.2
deepseek-v3.2Deepseek
88.0
2.2K
128K
¥2.09 / ¥3.1输入/输出
43
Ddeepseek-v4-flash
deepseek-v4-flashDeepseek
87.7
957
1M
¥1.01 / ¥2.02输入/输出
44
Ogpt-5.2-high
gpt-5.2-highOpenai
87.4
2.3K
400K
¥12.6 / ¥101输入/输出
45
TE
hunyuan-hy3-preview
hunyuan-hy3-previewTencent
87.1
360
256K
¥0 / ¥0输入/输出
46
M
longcat-flash-chat
longcat-flash-chatMeituan
86.8
545
128K
¥1.08 / ¥10.8输入/输出
47
Mkimi-k2.5-instant
kimi-k2.5-instantMoonshot
86.5
333
262K
¥4.32 / ¥21.6输入/输出
48
Mkimi-k2-thinking-turbo
kimi-k2-thinking-turboMoonshot
86.2
2.9K
262K
¥17.3 / ¥72输入/输出
49
Aqwen3-235b-a22b-instruct-2507
qwen3-235b-a22b-instruct-2507Alibaba
85.9
4.8K
128K
¥2.09 / ¥8.23输入/输出
50
Aqwen3.5-27b
qwen3.5-27bAlibaba
85.6
1.4K
262K
¥2.16 / ¥17.3输入/输出
51
Ggemini-3-flash (thinking-minimal)
gemini-3-flash (thinking-minimal)Google
85.3
2.6K
1.05M
¥3.6 / ¥21.6输入/输出
52
Aqwen3-vl-235b-a22b-thinking
qwen3-vl-235b-a22b-thinkingAlibaba
85.0
379
131K
¥2.06 / ¥8.26输入/输出
53
Aqwen3-235b-a22b-thinking-2507
qwen3-235b-a22b-thinking-2507Alibaba
84.8
492
131K
¥2.07 / ¥8.26输入/输出
54
Ogpt-5.4-mini-high
gpt-5.4-mini-highOpenai
84.5
1.4K
400K
¥5.4 / ¥32.4输入/输出
55
Ddeepseek-v3.1-thinking
deepseek-v3.1-thinkingDeepseek
84.2
565
128K
¥1.44 / ¥5.04输入/输出
56
Mlongcat-flash-chat-2602-exp
longcat-flash-chat-2602-expMeituan
83.9
1.3K
128K
¥1.08 / ¥10.8输入/输出
57
Xgrok-4.20-beta1
grok-4.20-beta1Xai
83.6
1.3K
2M
¥14.4 / ¥43.2输入/输出
58
Aqwen3-next-80b-a3b-instruct
qwen3-next-80b-a3b-instructAlibaba
83.3
1.1K
131K
¥1.04 / ¥4.13输入/输出
59
Ddeepseek-v3.2-exp-thinking
deepseek-v3.2-exp-thinkingDeepseek
83.0
393
128K
¥0 / ¥0输入/输出
60
Xgrok-4-fast-chat
grok-4-fast-chatXai
82.7
320
2M
¥1.44 / ¥3.6输入/输出
61
Ddeepseek-v3.2-thinking
deepseek-v3.2-thinkingDeepseek
82.4
1.7K
128K
¥2.09 / ¥3.1输入/输出
62
Ogpt-5.2-chat-latest-20260210
gpt-5.2-chat-latest-20260210Openai
82.1
1.7K
400K
¥12.6 / ¥101输入/输出
63
B
dola-seed-2.0-pro
dola-seed-2.0-proBytedance
81.8
2K
-
-
64
M
minimax-m2.7
minimax-m2.7Minimax
81.5
1.2K
205K
¥0 / ¥0输入/输出
65
Aclaude-opus-4-1-20250805-thinking-16k
claude-opus-4-1-20250805-thinking-16kAnthropic
81.2
2.4K
200K
¥108 / ¥540输入/输出
66
Aqwen3.5-122b-a10b
qwen3.5-122b-a10bAlibaba
80.9
1.4K
262K
¥2.88 / ¥23输入/输出
67
Ddeepseek-v4-flash-thinking
deepseek-v4-flash-thinkingDeepseek
80.6
886
1M
¥1.01 / ¥2.02输入/输出
68
Bernie-5.0-0110
ernie-5.0-0110Baidu
80.4
1.6K
128K
¥7.92 / ¥14.4输入/输出
69
Bernie-5.0-preview-1022
ernie-5.0-preview-1022Baidu
80.1
267
128K
¥7.92 / ¥14.4输入/输出
70
A
amazon-nova-experimental-chat-11-10
amazon-nova-experimental-chat-11-10Amazon
79.8
1.1K
-
-
71
Zglm-4.5
glm-4.5Zai
79.5
1.3K
131K
¥4.32 / ¥15.8输入/输出
72
Aclaude-opus-4-1-20250805
claude-opus-4-1-20250805Anthropic
79.2
3.8K
200K
¥108 / ¥540输入/输出
73
Xgrok-4-0709
grok-4-0709Xai
78.9
2.2K
256K
¥21.6 / ¥108输入/输出
74
MA
mistral-large-3
mistral-large-3Mistral
78.6
1.9K
262K
¥3.6 / ¥10.8输入/输出
75
Ogpt-5.4-nano-high
gpt-5.4-nano-highOpenai
78.3
1.4K
400K
¥1.44 / ¥9输入/输出
76
Oo3-2025-04-16
o3-2025-04-16Openai
78.0
3.5K
200K
¥14.4 / ¥57.6输入/输出
77
Ggemini-2.5-flash
gemini-2.5-flashGoogle
77.7
6.7K
1.05M
¥2.16 / ¥18输入/输出
78
Ggemini-2.5-flash-preview-09-2025
gemini-2.5-flash-preview-09-2025Google
77.4
1.6K
1M
¥2.16 / ¥18输入/输出
79
Ogpt-5.1
gpt-5.1Openai
77.1
2K
400K
¥9 / ¥72输入/输出
80
Aqwen3-vl-235b-a22b-instruct
qwen3-vl-235b-a22b-instructAlibaba
76.8
591
128K
¥2.16 / ¥8.64输入/输出
81
Mminimax-m2.1-preview
minimax-m2.1-previewMinimax
76.5
652
205K
¥0 / ¥0输入/输出
82
Ggemini-3.1-flash-lite-preview
gemini-3.1-flash-lite-previewGoogle
76.2
1.9K
1.05M
¥1.8 / ¥10.8输入/输出
83
Aqwen3-32b
qwen3-32bAlibaba
76.0
282
131K
¥2.07 / ¥8.26输入/输出
84
Aclaude-haiku-4-5-20251001
claude-haiku-4-5-20251001Anthropic
75.7
3.8K
200K
¥7.2 / ¥36输入/输出
85
Ddeepseek-r1-0528
deepseek-r1-0528Deepseek
75.4
949
164K
¥3.6 / ¥15.5输入/输出
86
Xgrok-4.3
grok-4.3Xai
75.1
813
1M
¥9 / ¥18输入/输出
87
Ogpt-5.2
gpt-5.2Openai
74.8
2.3K
400K
¥12.6 / ¥101输入/输出
88
Ogpt-5.5-instant
gpt-5.5-instantOpenai
74.5
1.5K
400K
¥9 / ¥72输入/输出
89
ST
step-3.5-flash
step-3.5-flashStepfun
74.2
1.7K
256K
¥0.69 / ¥2.07输入/输出
90
MAmistral-medium-2508
mistral-medium-2508Mistral
73.9
4.7K
262K
¥2.88 / ¥14.4输入/输出
91
Ddeepseek-v3.1
deepseek-v3.1Deepseek
73.6
839
128K
¥1.44 / ¥5.04输入/输出
92
MImimo-v2-flash (non-thinking)
mimo-v2-flash (non-thinking)Xiaomi
73.3
2.1K
262K
¥0.72 / ¥2.16输入/输出
93
Ddeepseek-v3.2-exp
deepseek-v3.2-expDeepseek
73.0
634
128K
¥0 / ¥0输入/输出
94
Aamazon-nova-experimental-chat-10-20
amazon-nova-experimental-chat-10-20Amazon
72.7
594
-
-
95
Xgrok-4.1-thinking
grok-4.1-thinkingXai
72.4
2.9K
200K
¥14.4 / ¥72输入/输出
96
Aqwen3-max-2025-09-23
qwen3-max-2025-09-23Alibaba
72.1
478
258K
¥6.19 / ¥24.7输入/输出
97
Aqwen3.5-flash
qwen3.5-flashAlibaba
71.8
1.5K
1M
¥1.24 / ¥12.4输入/输出
98
Zglm-4.7
glm-4.7Zai
71.6
541
205K
¥0 / ¥0输入/输出
99
Ogpt-4.5-preview-2025-02-27
gpt-4.5-preview-2025-02-27Openai
71.3
1.3K
8.19K
¥216 / ¥432输入/输出
100
Aqwen3-30b-a3b-instruct-2507
qwen3-30b-a3b-instruct-2507Alibaba
71.0
1.2K
262K
¥2.16 / ¥3.6输入/输出
101
Bernie-5.0-preview-1203
ernie-5.0-preview-1203Baidu
70.7
477
128K
¥7.92 / ¥14.4输入/输出
102
Xgrok-4.1
grok-4.1Xai
70.4
3.2K
200K
¥14.4 / ¥72输入/输出
103
Aqwen3.5-35b-a3b
qwen3.5-35b-a3bAlibaba
70.1
1.4K
262K
¥1.8 / ¥14.4输入/输出
104
Ochatgpt-4o-latest-20250326
chatgpt-4o-latest-20250326Openai
69.8
4.8K
128K
¥18 / ¥72输入/输出
105
Xgrok-3-preview-02-24
grok-3-preview-02-24Xai
69.5
2.4K
1M
¥9 / ¥18输入/输出
106
Ogpt-5-high
gpt-5-highOpenai
69.2
1.7K
400K
¥9 / ¥72输入/输出
107
Xgrok-4-1-fast-reasoning
grok-4-1-fast-reasoningXai
68.9
2.6K
2M
¥1.44 / ¥3.6输入/输出
108
Aqwen3-235b-a22b-no-thinking
qwen3-235b-a22b-no-thinkingAlibaba
68.6
2.3K
131K
¥2.07 / ¥8.26输入/输出
109
NV
nvidia-nemotron-3-super-120b-a12b
nvidia-nemotron-3-super-120b-a12bNvidia
68.3
414
262K
¥1.44 / ¥5.76输入/输出
110
NVnvidia-llama-3.3-nemotron-super-49b-v1.5
nvidia-llama-3.3-nemotron-super-49b-v1.5Nvidia
68.0
183
131K
¥2.88 / ¥2.88输入/输出
111
Xgrok-4-fast-reasoning
grok-4-fast-reasoningXai
67.7
941
2M
¥1.44 / ¥3.6输入/输出
112
Oo3-mini-high
o3-mini-highOpenai
67.4
1.7K
200K
¥7.92 / ¥31.7输入/输出
113
Mkimi-k2-0905-preview
kimi-k2-0905-previewMoonshot
67.2
641
262K
¥4.32 / ¥18输入/输出
114
Ogpt-5-chat
gpt-5-chatOpenai
66.9
1.6K
400K
¥9 / ¥72输入/输出
115
Aqwen3-235b-a22b
qwen3-235b-a22bAlibaba
66.6
1.5K
131K
¥2.07 / ¥8.26输入/输出
116
Zglm-4.5-air
glm-4.5-airZai
66.3
1.5K
131K
¥0 / ¥0输入/输出
117
Aclaude-opus-4-20250514-thinking-16k
claude-opus-4-20250514-thinking-16kAnthropic
66.0
2K
200K
¥108 / ¥540输入/输出
118
Oo1-2024-12-17
o1-2024-12-17Openai
65.7
2.6K
128K
¥108 / ¥432输入/输出
119
MImimo-v2-flash (thinking)
mimo-v2-flash (thinking)Xiaomi
65.4
426
262K
¥0.72 / ¥2.16输入/输出
120
Ddeepseek-r1
deepseek-r1Deepseek
65.1
1.5K
164K
¥5.04 / ¥18输入/输出
121
TEhunyuan-t1-20250711
hunyuan-t1-20250711Tencent
64.8
225
131K
¥0 / ¥0输入/输出
122
Aqwen3-next-80b-a3b-thinking
qwen3-next-80b-a3b-thinkingAlibaba
64.5
671
131K
¥1.04 / ¥10.3输入/输出
123
Ogpt-oss-120b
gpt-oss-120bOpenai
64.2
1.6K
131K
¥1.08 / ¥4.32输入/输出
124
Ogpt-5.3-chat-latest
gpt-5.3-chat-latestOpenai
63.9
1.7K
128K
¥12.6 / ¥101输入/输出
125
Ogpt-5-mini-high
gpt-5-mini-highOpenai
63.6
1.3K
400K
¥1.8 / ¥14.4输入/输出
126
Mminimax-m2.5
minimax-m2.5Minimax
63.3
1.9K
205K
¥0 / ¥0输入/输出
127
Oo1-preview
o1-previewOpenai
63.0
3.9K
128K
¥108 / ¥432输入/输出
128
Aclaude-sonnet-4-20250514-thinking-32k
claude-sonnet-4-20250514-thinking-32kAnthropic
62.8
1.9K
200K
¥21.6 / ¥108输入/输出
129
Oo3-mini
o3-miniOpenai
62.5
4.4K
200K
¥7.92 / ¥31.7输入/输出
130
UNintellect-3
-
62.2
233
131K
¥1.44 / ¥7.92输入/输出
131
NVnvidia-nemotron-3-nano-30b-a3b-bf16
nvidia-nemotron-3-nano-30b-a3b-bf16Nvidia
61.9
661
131K
¥0 / ¥0输入/输出
132
Aclaude-opus-4-20250514
claude-opus-4-20250514Anthropic
61.6
2.6K
200K
¥108 / ¥540输入/输出
133
Oo4-mini-2025-04-16
o4-mini-2025-04-16Openai
61.3
2.8K
200K
¥7.92 / ¥31.7输入/输出
134
Zglm-4.7-flash
glm-4.7-flashZai
61.0
481
200K
¥0 / ¥0输入/输出
135
STstep-3
step-3Stepfun
60.7
306
65.5K
¥1.8 / ¥4.68输入/输出
136
Ggemini-2.5-flash-lite-preview-09-2025-no-thinking
gemini-2.5-flash-lite-preview-09-2025-no-thinkingGoogle
60.4
2.1K
1.05M
¥0.72 / ¥2.88输入/输出
137
AG
ling-flash-2.0
ling-flash-2.0Ant Group
60.1
371
131K
¥1.01 / ¥4.1输入/输出
138
Anova-2-lite
nova-2-liteAmazon
59.8
583
128K
¥2.38 / ¥19.8输入/输出
139
Mminimax-m2
minimax-m2Minimax
59.5
285
197K
¥0 / ¥0输入/输出
140
Aqwen2.5-max
qwen2.5-maxAlibaba
59.2
2.9K
32K
¥11.5 / ¥46输入/输出
141
Ggemini-2.5-flash-lite-preview-06-17-thinking
gemini-2.5-flash-lite-preview-06-17-thinkingGoogle
58.9
2K
65.5K
¥0.72 / ¥2.88输入/输出
142
Oo1-mini
o1-miniOpenai
58.7
6.3K
128K
¥7.92 / ¥31.7输入/输出
143
Aqwen3-coder-480b-a35b-instruct
qwen3-coder-480b-a35b-instructAlibaba
58.4
1.5K
262K
¥6.2 / ¥24.8输入/输出
144
Xgrok-3-mini-high
grok-3-mini-highXai
58.1
987
128K
¥0 / ¥0输入/输出
145
Ddeepseek-v3-0324
deepseek-v3-0324Deepseek
57.8
2.9K
75K
¥1.44 / ¥5.76输入/输出
146
Mminimax-m1
minimax-m1Minimax
57.5
1.8K
1M
¥0.95 / ¥9.03输入/输出
147
Aqwen3-30b-a3b
qwen3-30b-a3bAlibaba
57.2
1.6K
128K
¥0.79 / ¥7.78输入/输出
148
UNtrinity-large-thinking
-
56.9
1.3K
262K
¥1.8 / ¥6.48输入/输出
149
TEhunyuan-turbos-20250416
hunyuan-turbos-20250416Tencent
56.6
795
131K
¥0 / ¥0输入/输出
150
AGring-flash-2.0
ring-flash-2.0Ant Group
56.3
360
131K
¥1.01 / ¥4.1输入/输出
151
Ogpt-4.1-2025-04-14
gpt-4.1-2025-04-14Openai
56.0
3.1K
1.05M
¥14.4 / ¥57.6输入/输出
152
Xgrok-3-mini-beta
grok-3-mini-betaXai
55.7
1.5K
1M
¥9 / ¥18输入/输出
153
Mkimi-k2-0711-preview
kimi-k2-0711-previewMoonshot
55.4
1.6K
131K
¥4.32 / ¥18输入/输出
154
Ggemini-2.0-flash-001
gemini-2.0-flash-001Google
55.1
3.8K
1.05M
¥1.08 / ¥4.32输入/输出
155
Aclaude-sonnet-4-20250514
claude-sonnet-4-20250514Anthropic
54.8
2.3K
200K
¥21.6 / ¥108输入/输出
156
Aqwq-32b
qwq-32bAlibaba
54.5
1.6K
131K
¥2.07 / ¥6.2输入/输出
157
MAmistral-small-2506
mistral-small-2506Mistral
54.3
993
262K
¥2.88 / ¥14.4输入/输出
158
MAmistral-medium-2505
mistral-medium-2505Mistral
54.0
2.2K
262K
¥2.88 / ¥14.4输入/输出
159
UNtrinity-large-preview
-
53.7
1.5K
262K
¥1.8 / ¥6.48输入/输出
160
AI
olmo-3.1-32b-instruct
olmo-3.1-32b-instructAllenai
53.4
428
200K
¥14.4 / ¥57.6输入/输出
161
Ogpt-4.1-mini-2025-04-14
gpt-4.1-mini-2025-04-14Openai
53.1
2.5K
1.05M
¥2.88 / ¥11.5输入/输出
162
NVllama-3.3-nemotron-49b-super-v1
llama-3.3-nemotron-49b-super-v1Nvidia
52.8
207
131K
¥0 / ¥0输入/输出
163
STstep-1o-turbo-202506
step-1o-turbo-202506Stepfun
52.5
569
-
-
164
Aclaude-3-7-sonnet-20250219-thinking-32k
claude-3-7-sonnet-20250219-thinking-32kAnthropic
52.2
2.6K
-
-
165
Ggemma-3-12b-it
gemma-3-12b-itGoogle
51.9
333
128K
¥1.96 / ¥1.96输入/输出
166
IB
granite-4.1-8b
granite-4.1-8bIbm
51.6
187
131K
¥0.36 / ¥0.72输入/输出
167
Aqwen-plus-0125
qwen-plus-0125Alibaba
51.3
627
1M
¥0.83 / ¥2.07输入/输出
168
AIolmo-3-32b-think
olmo-3-32b-thinkAllenai
51.0
202
128K
¥2.16 / ¥3.24输入/输出
169
Zglm-4.5v
glm-4.5vZai
50.7
234
64K
¥4.32 / ¥13输入/输出
170
Ggemma-3-27b-it
gemma-3-27b-itGoogle
50.4
3.2K
128K
¥2.15 / ¥2.15输入/输出
171
Ogpt-5-nano-high
gpt-5-nano-highOpenai
50.1
439
400K
¥0.36 / ¥2.88输入/输出
172
Ggemini-1.5-pro-002
gemini-1.5-pro-002Google
49.9
6.5K
-
-
173
Ggemini-2.0-flash-lite-preview-02-05
gemini-2.0-flash-lite-preview-02-05Google
49.6
2.4K
1.05M
¥0.54 / ¥2.16输入/输出
174
Aclaude-3-7-sonnet-20250219
claude-3-7-sonnet-20250219Anthropic
49.3
3.2K
200K
¥21.6 / ¥108输入/输出
175
Ddeepseek-v3
deepseek-v3Deepseek
49.0
2.3K
128K
¥0 / ¥0输入/输出
176
CO
command-a-03-2025
command-a-03-2025Cohere
48.7
3.6K
256K
¥18 / ¥72输入/输出
177
STstep-2-16k-exp-202412
step-2-16k-exp-202412Stepfun
48.4
543
16.4K
¥37.5 / ¥118输入/输出
178
Aclaude-3-5-sonnet-20241022
claude-3-5-sonnet-20241022Anthropic
48.1
8.6K
200K
¥21.6 / ¥108输入/输出
179
UNyi-lightning
-
47.8
3.2K
12K
¥1.44 / ¥1.44输入/输出
180
UNathene-v2-chat
-
47.5
2.9K
-
-
181
Ogpt-oss-20b
gpt-oss-20bOpenai
47.2
583
131K
¥0.32 / ¥1.3输入/输出
182
Aqwen2.5-plus-1127
qwen2.5-plus-1127Alibaba
46.9
1.3K
-
-
183
TEhunyuan-large-2025-02-10
hunyuan-large-2025-02-10Tencent
46.6
431
-
-
184
AIolmo-3.1-32b-think
olmo-3.1-32b-thinkAllenai
46.3
340
200K
¥14.4 / ¥57.6输入/输出
185
TEhunyuan-turbo-0110
hunyuan-turbo-0110Tencent
46.0
216
-
-
186
TEhunyuan-large-vision
hunyuan-large-visionTencent
45.7
364
-
-
187
Mllama-4-maverick-17b-128e-instruct
llama-4-maverick-17b-128e-instructMeta
45.5
2.6K
1M
¥1.8 / ¥6.26输入/输出
188
Aclaude-3-5-sonnet-20240620
claude-3-5-sonnet-20240620Anthropic
45.2
10K
200K
¥21.6 / ¥108输入/输出
189
Zglm-4-plus-0111
glm-4-plus-0111Zai
44.9
638
128K
¥72 / ¥72输入/输出
190
Aqwen2.5-72b-instruct
qwen2.5-72b-instructAlibaba
44.6
4.6K
131K
¥4.13 / ¥12.4输入/输出
191
TEhunyuan-turbos-20250226
hunyuan-turbos-20250226Tencent
44.3
224
131K
¥0 / ¥0输入/输出
192
TEhunyuan-standard-2025-02-10
hunyuan-standard-2025-02-10Tencent
44.0
441
-
-
193
NVllama-3.1-nemotron-70b-instruct
llama-3.1-nemotron-70b-instructNvidia
43.7
874
128K
¥0 / ¥0输入/输出
194
Mllama-4-scout-17b-16e-instruct
llama-4-scout-17b-16e-instructMeta
43.4
1.8K
128K
¥1.44 / ¥5.62输入/输出
195
Ogpt-4o-2024-05-13
gpt-4o-2024-05-13Openai
43.1
13.3K
128K
¥36 / ¥108输入/输出
196
Xgrok-2-2024-08-13
grok-2-2024-08-13Xai
42.8
7.6K
1M
¥9 / ¥18输入/输出
197
Ogpt-4o-2024-08-06
gpt-4o-2024-08-06Openai
42.5
5.8K
128K
¥18 / ¥72输入/输出
198
Zglm-4-plus
glm-4-plusZai
42.2
3K
128K
¥54 / ¥54输入/输出
199
Aqwen-max-0919
qwen-max-0919Alibaba
41.9
1.8K
131K
¥2.48 / ¥9.91输入/输出
200
Ggemini-1.5-flash-002
gemini-1.5-flash-002Google
41.6
4.1K
2M
¥0.54 / ¥2.2输入/输出
201
Ddeepseek-v2.5-1210
deepseek-v2.5-1210Deepseek
41.3
893
1M
¥1.01 / ¥2.02输入/输出
202
Ggemini-1.5-pro-001
gemini-1.5-pro-001Google
41.1
9.3K
-
-
203
Mllama-3.1-405b-instruct-fp8
llama-3.1-405b-instruct-fp8Meta
40.8
7.2K
128K
¥0 / ¥0输入/输出
204
Ggemini-advanced-0514
gemini-advanced-0514Google
40.5
5.8K
-
-
205
Mllama-3.1-405b-instruct-bf16
llama-3.1-405b-instruct-bf16Meta
40.2
4.4K
128K
¥0 / ¥0输入/输出
206
IBibm-granite-h-small
ibm-granite-h-smallIbm
39.9
279
-
-
207
Ogpt-4.1-nano-2025-04-14
gpt-4.1-nano-2025-04-14Openai
39.6
534
1.05M
¥14.4 / ¥57.6输入/输出
208
Ddeepseek-v2.5
deepseek-v2.5Deepseek
39.3
3K
1M
¥1.01 / ¥2.02输入/输出
209
Ggemma-3n-e4b-it
gemma-3n-e4b-itGoogle
39.0
1.4K
128K
¥0 / ¥0输入/输出
210
Xgrok-2-mini-2024-08-13
grok-2-mini-2024-08-13Xai
38.7
6.3K
1M
¥9 / ¥18输入/输出
211
MAmistral-small-3.1-24b-instruct-2503
mistral-small-3.1-24b-instruct-2503Mistral
38.4
1.9K
262K
¥2.88 / ¥14.4输入/输出
212
Ogpt-4o-mini-2024-07-18
gpt-4o-mini-2024-07-18Openai
38.1
8K
128K
¥1.08 / ¥4.32输入/输出
213
Ogpt-4-1106-preview
gpt-4-1106-previewOpenai
37.8
11.5K
8.19K
¥216 / ¥432输入/输出
214
Ogpt-4-turbo-2024-04-09
gpt-4-turbo-2024-04-09Openai
37.5
11.8K
128K
¥72 / ¥216输入/输出
215
Mllama-3.3-70b-instruct
llama-3.3-70b-instructMeta
37.2
5.2K
128K
¥0 / ¥0输入/输出
216
MAmistral-large-2407
mistral-large-2407Mistral
37.0
5.7K
131K
¥14.4 / ¥43.2输入/输出
217
Aclaude-3-opus-20240229
claude-3-opus-20240229Anthropic
36.7
23K
200K
¥108 / ¥540输入/输出
218
Aamazon-nova-pro-v1.0
amazon-nova-pro-v1.0Amazon
36.4
2.7K
300K
¥5.76 / ¥23输入/输出
219
Ogpt-4-0125-preview
gpt-4-0125-previewOpenai
36.1
10.8K
8.19K
¥216 / ¥432输入/输出
220
MAmistral-large-2411
mistral-large-2411Mistral
35.8
3K
128K
¥14.4 / ¥43.2输入/输出
221
Ggemma-3-4b-it
gemma-3-4b-itGoogle
35.5
365
128K
¥1.44 / ¥1.44输入/输出
222
MAmagistral-medium-2506
magistral-medium-2506Mistral
35.2
582
128K
¥14.4 / ¥36输入/输出
223
M
phi-4
phi-4Microsoft
34.9
2.4K
128K
¥0.9 / ¥3.6输入/输出
224
Mllama-3.1-70b-instruct
llama-3.1-70b-instructMeta
34.6
6.5K
131K
¥2.88 / ¥2.88输入/输出
225
Aclaude-3-5-haiku-20241022
claude-3-5-haiku-20241022Anthropic
34.3
5.8K
200K
¥5.76 / ¥28.8输入/输出
226
AIllama-3.1-tulu-3-70b
llama-3.1-tulu-3-70bAllenai
34.0
349
-
-
227
TEhunyuan-standard-256k
hunyuan-standard-256kTencent
33.7
292
-
-
228
Aqwen2.5-coder-32b-instruct
qwen2.5-coder-32b-instructAlibaba
33.4
619
131K
¥2.07 / ¥6.2输入/输出
229
MAmistral-small-24b-instruct-2501
mistral-small-24b-instruct-2501Mistral
33.1
1.4K
262K
¥2.88 / ¥14.4输入/输出
230
UNreka-core-20240904
-
32.8
1K
-
-
231
Ggemini-1.5-flash-001
gemini-1.5-flash-001Google
32.6
7.3K
2M
¥0.54 / ¥2.2输入/输出
232
Aamazon-nova-lite-v1.0
amazon-nova-lite-v1.0Amazon
32.3
2.2K
300K
¥0.43 / ¥1.73输入/输出
233
UNathene-70b-0725
-
32.0
2.6K
-
-
234
Ddeepseek-coder-v2
deepseek-coder-v2Deepseek
31.7
1.7K
1M
¥1.01 / ¥2.02输入/输出
235
Aqwen2-72b-instruct
qwen2-72b-instructAlibaba
31.4
4.2K
131K
¥4.13 / ¥12.4输入/输出
236
Zglm-4-0520
glm-4-0520Zai
31.1
1.1K
128K
¥108 / ¥108输入/输出
237
NVllama-3.1-nemotron-51b-instruct
llama-3.1-nemotron-51b-instructNvidia
30.8
422
128K
¥0 / ¥0输入/输出
238
Ogpt-4-0314
gpt-4-0314Openai
30.5
6.1K
8.19K
¥216 / ¥432输入/输出
239
UNjamba-1.5-large
-
30.2
962
256K
¥0 / ¥0输入/输出
240
Ggemini-1.5-flash-8b-001
gemini-1.5-flash-8b-001Google
29.9
4.2K
2M
¥0.54 / ¥2.2输入/输出
241
Aqwq-32b-preview
qwq-32b-previewAlibaba
29.6
422
131K
¥2.07 / ¥6.2输入/输出
242
Ggemma-2-27b-it
gemma-2-27b-itGoogle
29.3
8.9K
8.19K
¥0.58 / ¥0.58输入/输出
243
NVnemotron-4-340b-instruct
nemotron-4-340b-instructNvidia
29.0
2.1K
-
-
244
Mllama-3-70b-instruct
llama-3-70b-instructMeta
28.7
18.7K
8.19K
¥3.67 / ¥5.33输入/输出
245
Aclaude-3-sonnet-20240229
claude-3-sonnet-20240229Anthropic
28.4
12.4K
200K
¥21.6 / ¥108输入/输出
246
Aamazon-nova-micro-v1.0
amazon-nova-micro-v1.0Amazon
28.2
2K
128K
¥0.25 / ¥1.01输入/输出
247
COc4ai-aya-expanse-32b
c4ai-aya-expanse-32bCohere
27.9
3.2K
-
-
248
UNreka-flash-20240904
-
27.6
1.1K
65.5K
¥0.72 / ¥1.44输入/输出
249
Ogpt-4-0613
gpt-4-0613Openai
27.3
9.9K
8.19K
¥216 / ¥432输入/输出
250
COcommand-r-plus-08-2024
command-r-plus-08-2024Cohere
27.0
1.2K
128K
¥18 / ¥72输入/输出
251
UNgemma-2-9b-it-simpo
-
26.7
1.1K
8.19K
¥1.44 / ¥1.44输入/输出
252
AIolmo-2-0325-32b-instruct
olmo-2-0325-32b-instructAllenai
26.4
336
-
-
253
Ggemma-2-9b-it
gemma-2-9b-itGoogle
26.1
6.3K
8.19K
¥1.44 / ¥1.44输入/输出
254
AIllama-3.1-tulu-3-8b
llama-3.1-tulu-3-8bAllenai
25.8
304
-
-
255
Aqwen1.5-110b-chat
qwen1.5-110b-chatAlibaba
25.5
2.9K
-
-
256
MAmistral-large-2402
mistral-large-2402Mistral
25.2
7K
262K
¥2.88 / ¥14.4输入/输出
257
Aclaude-3-haiku-20240307
claude-3-haiku-20240307Anthropic
24.9
13.5K
200K
¥1.8 / ¥9输入/输出
258
IBgranite-3.1-2b-instruct
granite-3.1-2b-instructIbm
24.6
341
-
-
259
UNyi-1.5-34b-chat
-
24.3
2.6K
-
-
260
MAministral-8b-2410
ministral-8b-2410Mistral
24.0
543
128K
¥0.72 / ¥0.72输入/输出
261
UNinternlm2_5-20b-chat
-
23.8
1.1K
-
-
262
IBgranite-3.1-8b-instruct
granite-3.1-8b-instructIbm
23.5
320
-
-
263
Mllama-3.1-8b-instruct
llama-3.1-8b-instructMeta
23.2
6K
131K
¥0.79 / ¥0.79输入/输出
264
MAmixtral-8x22b-instruct-v0.1
mixtral-8x22b-instruct-v0.1Mistral
22.9
6K
64K
¥14.4 / ¥43.2输入/输出
265
MAmistral-medium
mistral-mediumMistral
22.6
3.8K
262K
¥2.88 / ¥14.4输入/输出
266
COc4ai-aya-expanse-8b
c4ai-aya-expanse-8bCohere
22.3
1.1K
-
-
267
Mphi-3-medium-4k-instruct
phi-3-medium-4k-instructMicrosoft
22.0
2.7K
4.1K
¥1.22 / ¥4.9输入/输出
268
Aqwen1.5-72b-chat
qwen1.5-72b-chatAlibaba
21.7
4.5K
-
-
269
UNreka-flash-21b-20240226-online
-
21.4
1.9K
-
-
270
COcommand-r-08-2024
command-r-08-2024Cohere
21.1
1.4K
128K
¥18 / ¥72输入/输出
271
COcommand-r-plus
command-r-plusCohere
20.8
8.8K
128K
¥18 / ¥72输入/输出
272
Aqwen1.5-32b-chat
qwen1.5-32b-chatAlibaba
20.5
2.4K
-
-
273
UNjamba-1.5-mini
-
20.2
932
256K
¥0 / ¥0输入/输出
274
UNreka-flash-21b-20240226
-
19.9
3K
-
-
275
Mllama-3-8b-instruct
llama-3-8b-instructMeta
19.6
12.8K
8.19K
¥0.29 / ¥0.29输入/输出
276
Mphi-3-small-8k-instruct
phi-3-small-8k-instructMicrosoft
19.4
1.8K
8.19K
¥1.08 / ¥4.32输入/输出
277
IBgranite-3.0-8b-instruct
granite-3.0-8b-instructIbm
19.1
675
-
-
278
Mphi-3-mini-4k-instruct-june-2024
phi-3-mini-4k-instruct-june-2024Microsoft
18.8
1.4K
4.1K
¥0.94 / ¥3.74输入/输出
279
Ggemma-2-2b-it
gemma-2-2b-itGoogle
18.5
5.6K
128K
¥0 / ¥0输入/输出
280
UNzephyr-orpo-141b-A35b-v0.1
-
18.2
526
200K
¥108 / ¥432输入/输出
281
MAmixtral-8x7b-instruct-v0.1
mixtral-8x7b-instruct-v0.1Mistral
17.9
8.5K
32K
¥5.04 / ¥5.04输入/输出
282
Ggemini-pro
gemini-proGoogle
17.6
844
1.05M
¥14.4 / ¥86.4输入/输出
283
Aqwen1.5-14b-chat
qwen1.5-14b-chatAlibaba
17.3
2K
-
-
284
UNdbrx-instruct-preview
-
17.0
3.6K
-
-
285
IBgranite-3.0-2b-instruct
granite-3.0-2b-instructIbm
16.7
751
-
-
286
UNstarling-lm-7b-beta
-
16.4
1.8K
200K
¥5.4 / ¥18.7输入/输出
287
Mllama-3.2-3b-instruct
llama-3.2-3b-instructMeta
16.1
980
131K
¥0.22 / ¥0.35输入/输出
288
Ogpt-3.5-turbo-1106
gpt-3.5-turbo-1106Openai
15.8
1.7K
16.4K
¥7.2 / ¥14.4输入/输出
289
Ogpt-3.5-turbo-0125
gpt-3.5-turbo-0125Openai
15.5
7.5K
16.4K
¥3.6 / ¥10.8输入/输出
290
COcommand-r
command-rCohere
15.2
5.9K
128K
¥18 / ¥72输入/输出
291
Ggemini-pro-dev-api
gemini-pro-dev-apiGoogle
15.0
2K
1.05M
¥14.4 / ¥86.4输入/输出
292
UNsmollm2-1.7b-instruct
-
14.7
240
-
-
293
UNyi-34b-chat
-
14.4
1.7K
-
-
294
Mphi-3-mini-4k-instruct
phi-3-mini-4k-instructMicrosoft
14.1
2.2K
4.1K
¥0.94 / ¥3.74输入/输出
295
Aqwen1.5-7b-chat
qwen1.5-7b-chatAlibaba
13.8
578
-
-
296
Mwizardlm-70b
wizardlm-70bMicrosoft
13.5
764
-
-
297
UNtulu-2-dpo-70b
-
13.2
751
-
-
298
Mllama-3.2-1b-instruct
llama-3.2-1b-instructMeta
12.9
976
16.4K
¥0.07 / ¥0.08输入/输出
299
Ggemma-1.1-7b-it
gemma-1.1-7b-itGoogle
12.6
2.7K
-
-
300
UNsnowflake-arctic-instruct
-
12.3
4.2K
-
-
301
UNopenhermes-2.5-mistral-7b
-
12.0
568
1M
¥36 / ¥180输入/输出
302
UNopenchat-3.5-0106
-
11.7
1.5K
-
-
303
Ddeepseek-llm-67b-chat
deepseek-llm-67b-chatDeepseek
11.4
484
1M
¥1.01 / ¥2.02输入/输出
304
Mllama-2-70b-chat
llama-2-70b-chatMeta
11.1
4.1K
-
-
305
MAmistral-7b-instruct-v0.2
mistral-7b-instruct-v0.2Mistral
10.9
2.2K
262K
¥2.88 / ¥14.4输入/输出
306
Mphi-3-mini-128k-instruct
phi-3-mini-128k-instructMicrosoft
10.6
2.6K
128K
¥0.94 / ¥3.74输入/输出
307
UNstarling-lm-7b-alpha
-
10.3
1.1K
200K
¥5.4 / ¥18.7输入/输出
308
UNvicuna-33b
-
10.0
2.4K
-
-
309
Aqwen-14b-chat
qwen-14b-chatAlibaba
9.7
467
32.8K
¥1.04 / ¥3.1输入/输出
310
UNopenchat-3.5
-
9.4
826
-
-
311
Mllama-2-13b-chat
llama-2-13b-chatMeta
9.1
2K
-
-
312
NVllama2-70b-steerlm-chat
llama2-70b-steerlm-chatNvidia
8.8
353
-
-
313
Ggemma-7b-it
gemma-7b-itGoogle
8.5
982
-
-
314
UNsolar-10.7b-instruct-v1.0
-
8.2
529
128K
¥0 / ¥0输入/输出
315
UNdolphin-2.2.1-mistral-7b
-
7.9
201
262K
¥2.88 / ¥14.4输入/输出
316
Mcodellama-34b-instruct
codellama-34b-instructMeta
7.6
667
-
-
317
UNmpt-30b-chat
-
7.3
218
-
-
318
UNnous-hermes-2-mixtral-8x7b-dpo
-
7.0
464
1M
¥36 / ¥180输入/输出
319
UNzephyr-7b-beta
-
6.7
1.1K
-
-
320
Gpalm-2
palm-2Google
6.5
801
-
-
321
Ggemma-1.1-2b-it
gemma-1.1-2b-itGoogle
6.2
1.2K
-
-
322
Mllama-2-7b-chat
llama-2-7b-chatMeta
5.9
1.4K
128K
¥4.03 / ¥48输入/输出
323
UNvicuna-13b
-
5.6
1.9K
-
-
324
UNstripedhyena-nous-7b
-
5.3
530
-
-
325
UNguanaco-33b
-
5.0
236
200K
¥14.4 / ¥57.6输入/输出
326
AIolmo-7b-instruct
olmo-7b-instructAllenai
4.7
726
-
-
327
Mwizardlm-13b
wizardlm-13bMicrosoft
4.4
567
-
-
328
MAmistral-7b-instruct
mistral-7b-instructMistral
4.1
814
262K
¥2.88 / ¥14.4输入/输出
329
Ggemma-2b-it
gemma-2b-itGoogle
3.8
525
-
-
330
Aqwen1.5-4b-chat
qwen1.5-4b-chatAlibaba
3.5
820
-
-
331
UNvicuna-7b
-
3.2
590
-
-
332
UNchatglm3-6b
-
2.9
466
200K
¥5.4 / ¥18.7输入/输出
333
UNkoala-13b
-
2.6
692
-
-
334
UNRWKV-4-Raven-14B
-
2.3
488
-
-
335
UNchatglm-6b
-
2.1
466
200K
¥5.4 / ¥18.7输入/输出
336
UNmpt-7b-chat
-
1.8
397
-
-
337
UNoasst-pythia-12b
-
1.5
629
-
-
338
UNalpaca-13b
-
1.2
565
-
-
339
UNfastchat-t5-3b
-
0.9
400
-
-
340
UNdolly-v2-12b
-
0.6
343
-
-
341
UNstablelm-tuned-alpha-7b
-
0.3
294
-
-
342
Mllama-13b
llama-13bMeta
0.0
228
-
-
claude-opus-4-6 为什么排第一
claude-opus-4-6 当前排名第 1，厂商为 anthropic，百分制分数 100.0，样本量 1.9K。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
数学行业任务常见问题
数学行业任务排行榜看什么指标？
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。
数学行业任务模型应该怎么选？
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。