排名 模型
🥇
gemini-3-pro Google · Proprietary
1353±19.3
🥈
gemini-3-flash Google · Proprietary
1346±20.4
🥉
gemini-3-flash (thinking-minimal) Google · Proprietary
1310±25.5
4
gemini-2.5-flash-preview-09-2025 Google · Proprietary
1291±30.2
5
gpt-5.1 OpenAI · Proprietary
1281±23.7
6
gemini-2.5-pro Google · Proprietary
1268±13.1
7
gpt-5.1-high OpenAI · Proprietary
1267±23.7
8
chatgpt-4o-latest-20250326 OpenAI · Proprietary
1266±16.6
9
gpt-5.2 OpenAI · Proprietary
1266±39.2
10
kimi-k2.5-thinking 月之暗面 · Modified MIT
1262±37.6
11
claude-opus-4-20250514 Anthropic · Proprietary
1256±30.5
12
gemini-2.5-flash Google · Proprietary
1254±14.0
13
ernie-5.0-preview-1220 百度 · Proprietary
1254±29.3
14
grok-4-0709 xAI · Proprietary
1243±17.7
15
kimi-k2.5-instant 月之暗面 · Modified MIT
1233±43.7
16
qwen3-vl-235b-a22b-instruct 阿里巴巴 · Apache 2.0
1232±20.4
17
gpt-5-chat OpenAI · Proprietary
1227±17.0
18
gpt-5.2-high OpenAI · Proprietary
1226±38.3
19
gpt-4.1-2025-04-14 OpenAI · Proprietary
1222±17.0
20
o3-2025-04-16 OpenAI · Proprietary
1217±15.4
21
claude-3-7-sonnet-20250219-thinking-32k Anthropic · Proprietary
1213±37.5
22
gemini-2.5-flash-lite-preview-09-2025-no-thinking Google · Proprietary
1206±30.1
23
mistral-medium-2508 Mistral · Proprietary
1203±16.0
24
gemini-2.0-flash-001 Google · Proprietary
1203±32.8
25
claude-sonnet-4-20250514 Anthropic · Proprietary
1200±36.5
26
gpt-5-high OpenAI · Proprietary
1197±16.3
27
gpt-5-mini-high OpenAI · Proprietary
1196±19.7
28
mistral-small-2506 Mistral · Apache 2.0
1193±26.9
29
qwen3-vl-235b-a22b-thinking 阿里巴巴 · Apache 2.0
1191±44.3
30
gpt-4.1-mini-2025-04-14 OpenAI · Proprietary
1190±17.9
31
gemini-2.5-flash-lite-preview-06-17-thinking Google · Proprietary
1188±18.0
32
gemma-3-27b-it Google · Gemma
1187±23.2
33
o4-mini-2025-04-16 OpenAI · Proprietary
1179±17.1
34
mistral-medium-2505 Mistral · Proprietary
1166±24.0
35
glm-4.6v 智谱 ZAI · MIT
1154±40.1
36
llama-4-scout-17b-16e-instruct Meta · Llama
1141±40.1
37
mistral-small-3.1-24b-instruct-2503 Mistral · Apache 2.0
1137±21.4
38
llama-4-maverick-17b-128e-instruct Meta · Llama 4
1117±40.1

没有找到相关模型