AI Model Rankings
基于模型热度、真实调用趋势与能力指标,追踪主流 AI 模型的市场表现。
Top Models
主流模型每周使用趋势
LLM Leaderboard
按时间范围查询不同模型的调用量
1.
DeepSeek V4.1 Flash (batch)
33.6T tokens
↑48%
2.
space-bunny-alpha
28.0T tokens
↓1%
3.
GLM 5.3 Flash (batch)
10.5T tokens
↓1%
4.
MiMo-V2.6-Flash
10.1T tokens
↑11%
5.
Hy4 preview
6.74T tokens
↓10%
6.
GPT-6 Luna (batch)
6.46T tokens
↑28%
7.
nemotron-3-ultra-550b-a55b-20260604:free
5.95T tokens
↓2%
8.
DeepSeek V4 Flash 0731
5.53T tokens
↓21%
9.
Claude Opus 5.5 (batch)
3.37T tokens
↑74%
10.
jev-1.13-20260917
3.32T tokens
↑12%
11.
DeepSeek V4 Flash 0423
3.21T tokens
12.
GLM 5.3 (batch)
3.16T tokens
↑15%
13.
Gemini 3.8 Flash (batch)
2.13T tokens
↑1%
14.
GPT-5.6 Luna (batch)
1.95T tokens
↓75%
15.
GPT-5.6 Sol (batch)
1.77T tokens
↑15%
16.
Kimi K3 (batch)
1.72T tokens
↑10%
17.
Hy3
1.58T tokens
↓34%
18.
GLM 5.2
1.58T tokens
↑21%
19.
Muse Spark 1.3 Contributor
1.46T tokens
↓2%
20.
GPT-6.1 Sol
1.40T tokens
↑10%
Tool Calls
对比不同模型的工具调用使用量
1.
Others
265M calls
↑19%
2.
DeepSeek V4.1 Flash (batch)
168M calls
↑80%
3.
GLM 5.3 Flash (batch)
62M calls
↑17%
4.
MiMo-V2.6-Flash
42M calls
↑7%
5.
GPT-6 Luna (batch)
41M calls
↑21%
6.
space-bunny-alpha
33M calls
↓79%
7.
Hy4 preview
31M calls
↑30%
8.
GPT-5.6 Luna (batch)
30M calls
↓38%
9.
DeepSeek V4 Flash 0731
24M calls
↓8%
10.
DeepSeek V4 Flash 0423
20M calls
↑22%
Benchmarks
按综合能力指标对比模型表现
1.
GPT-5.5 (xhigh)
66.8
2.
Claude Opus 4.7 (Adaptive)
64.9
3.
MiMo-V2.5-Pro
63.7
4.
Grok 4.3
61.4
5.
Claude Sonnet 4.6
55.1
6.
Qwen3.6 35B A3B (Reasoning)
51.7
7.
MiniMax-M2.1
48.2
8.
Mistral Medium 3.5
45.6
9.
Grok 4.1 Fast (Reasoning)
43.1
10.
Gemini 3 Flash Preview
40.5
Fastest models
对比不同服务商下的模型吞吐表现
Highest throughput
1.
gpt-oss-safeguard-20b
645 tok/s
$0.07/M
2.
gpt-oss-20b
634 tok/s
$0.07/M
3.
gpt-oss-120b
626 tok/s
$0.35/M
4.
Mercury 2
355 tok/s
$0.25/M
5.
Qwen3 32B
351 tok/s
$0.29/M
6.
GLM 4.7
302 tok/s
$2.25/M
7.
MiniMax M2.5
261 tok/s
$0.30/M
8.
Llama 3.1 8B Instruct
230 tok/s
$0.10/M
9.
Qwen3.6 35B A3B
165 tok/s
$0.25/M
10.
Nano Banana (Gemini 2.5)
162 tok/s
$0.35/M
Context Length
按上下文窗口对比模型使用情况
10K
1.
Others
699M requests
↓7%
2.
jev-1.13-20260917
303M requests
↓0%
3.
GLM 5.3 Flash (batch)
276M requests
↑410%
4.
DeepSeek V4 Flash 0731
96M requests
↓16%
5.
DeepSeek V4 Flash 0423
96M requests
↓11%
6.
GPT-6 Luna (batch)
78M requests
↑15%
7.
DeepSeek V4.1 Flash (batch)
67M requests
↓6%
8.
Gemini 2.5 Flash Lite (batch)
56M requests
↓18%
9.
GPT-5.6 Luna (batch)
47M requests
↓23%
10.
gpt-oss-120b (batch)
41M requests
0%
Categories
按使用场景对比模型表现
Programming
1.
DeepSeek V4.1 Flash (batch)
11.3T tokens
↑117%
2.
Others
8.38T tokens
↑15%
3.
MiMo-V2.6-Flash
4.30T tokens
↑6%
4.
space-bunny-alpha
2.69T tokens
↓76%
5.
nemotron-3-ultra-550b-a55b-20260604:free
2.59T tokens
↑1%
6.
GLM 5.3 Flash (batch)
2.39T tokens
↑13%
7.
GPT-6 Luna (batch)
1.26T tokens
↑28%
8.
GLM 5.2
897B tokens
↑72%
9.
Hy4 preview
711B tokens
0%
10.
GLM 5.3 (batch)
672B tokens
↑10%
Languages
按自然语言使用量对比模型表现
1.
Others
6.87T tokens
↑0%
2.
DeepSeek V4.1 Flash (batch)
4.84T tokens
↑34%
3.
GLM 5.3 Flash (batch)
1.61T tokens
↓37%
4.
MiMo-V2.6-Flash
1.02T tokens
↑2%
5.
GPT-6 Luna (batch)
873B tokens
↓5%
6.
Hy4 preview
850B tokens
↑8%
7.
nemotron-3-ultra-550b-a55b-20260604:free
621B tokens
↑5%
8.
DeepSeek V4 Flash 0731
608B tokens
↓22%
9.
GLM 5.3 (batch)
414B tokens
0%
10.
GLM 5.2
406B tokens
0%
Programming
按编程语言使用量对比模型表现
Python
1.
Others
1.64T tokens
↓5%
2.
DeepSeek V4.1 Flash (batch)
1.08T tokens
↑43%
3.
GLM 5.3 Flash (batch)
423B tokens
↓40%
4.
MiMo-V2.6-Flash
214B tokens
↑2%
5.
GPT-6 Luna (batch)
203B tokens
↓8%
6.
GLM 5.2
184B tokens
↑45%
7.
Hy4 preview
147B tokens
↑52%
8.
DeepSeek V4 Flash 0731
135B tokens
↓22%
9.
DeepSeek V4 Flash 0423
98.6B tokens
↓12%
10.
nemotron-3-ultra-550b-a55b-20260604:free
97.1B tokens
0%
Images
模型处理图像任务的累计趋势
1.
GLM 5.3 Flash (batch)
1.3B requests
↑810%
2.
Others
278M requests
↓65%
3.
Gemini 2.5 Flash Lite (batch)
100M requests
↓54%
4.
DeepSeek V4.1 Flash (batch)
72M requests
↓48%
5.
GPT-6 Luna (batch)
51M requests
↓49%
6.
MiMo-V2.6-Flash
40M requests
↓53%
7.
Claude Opus 5.5 (batch)
32M requests
0%
8.
Gemini 3.8 Flash (batch)
30M requests
↓69%
9.
space-bunny-alpha
19M requests
↓91%
10.
Gemini 3.1 Flash Lite (batch)
19M requests
0%
Audio Input
模型处理音频输入的累计趋势
1.
GPT-4o Transcribe
26.4M prompts
↑16%
2.
Gemini 2.5 Flash
21.9M prompts
↓1%
3.
Whisper Large V3
18.2M prompts
↓4%
4.
MiniMax Speech 2.5
14.7M prompts
↑12%
5.
Gemini 3 Flash Preview
12.6M prompts
↑7%
6.
Claude Sonnet 4.6
10.1M prompts
↑3%
7.
Qwen Audio
8.8M prompts
↑9%
8.
Mistral Voxtral
7.4M prompts
2%
9.
GLM Audio
5.9M prompts
↑5%
10.
Llama 3.1 Audio
4.8M prompts
↓6%
Top Apps
按应用与 Agent 场景观察模型采用情况
1.
Hermes Agent
353B tokens
2.
OpenClaw
195B tokens
3.
Kilo Code
166B tokens
4.
Claude Code
70.5B tokens
5.
CSS AI Pro
66.7B tokens
6.
Descript
62.7B tokens
7.
pi
39.9B tokens
8.
Janitor AI
27.7B tokens
9.
ISEKAI ZERO
25B tokens
10.
Roo Code
22.8B tokens