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 Flash 0731
11.3T tokens
↑12%
2.
Hy3
9.83T tokens
↑4%
3.
GPT-5.6 Luna (batch)
5.80T tokens
↑19%
4.
MiMo-V2.5
5.46T tokens
↑9%
5.
DeepSeek V4 Flash 0423
4.78T tokens
↓13%
6.
GLM 5.2 (free)
4.34T tokens
↑18%
7.
Gemini 3.6 Flash (batch)
2.75T tokens
↑12%
8.
nemotron-3-ultra-550b-a55b-20260604:free
2.69T tokens
↑20%
9.
Claude Opus 5 (batch)
2.68T tokens
↑89%
10.
DeepSeek V4 Pro 0423
2.57T tokens
↑1%
11.
MiniMax M3 (batch)
1.67T tokens
↑2%
12.
laguna-s-2.1-20260720:free
1.67T tokens
↓5%
13.
Kimi K3
1.32T tokens
↓9%
14.
Claude Sonnet 5 (batch)
1.08T tokens
↑5%
15.
DeepSeek V4 Pro 0813
946B tokens
16.
GPT-5.6 Terra (batch)
943B tokens
↑12%
17.
Step 3.7 Flash
857B tokens
↓30%
18.
nemotron-3.5-lightning-20260807:free
836B tokens
↑39%
19.
Gemini 3.7 Flash (batch)
830B tokens
20.
Gemini 3 Flash Preview (batch)
821B tokens
↓8%
Tool Calls
对比不同模型的工具调用使用量
1.
Others
97M calls
↓9%
2.
Hy3
37M calls
↑3%
3.
DeepSeek V4 Flash 0731
36M calls
↓5%
4.
MiMo-V2.5
31M calls
↑133%
5.
GPT-5.6 Luna (batch)
27M calls
↑17%
6.
DeepSeek V4 Flash 0423
13M calls
↓17%
7.
GLM 5.2 (free)
12M calls
↓28%
8.
nemotron-3-ultra-550b-a55b-20260604:free
7M calls
0%
9.
Ling-3.0-flash
6M calls
↑7%
10.
DeepSeek V4 Pro 0423
5M calls
↓43%
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
321M requests
↑1%
2.
DeepSeek V4 Flash 0423
113M requests
↓5%
3.
DeepSeek V4 Flash 0731
63M requests
↑23%
4.
Gemini 2.5 Flash Lite (batch)
45M requests
↓7%
5.
gpt-oss-120b
32M requests
↑38%
6.
GPT-5.6 Luna (batch)
28M requests
↓28%
7.
Gemini 2.5 Flash (batch)
22M requests
↓9%
8.
Mistral Nemo
19M requests
0%
9.
Gemini 3 Flash Preview (batch)
18M requests
↓11%
10.
Gemma 4 26B A4B (free)
18M requests
0%
Categories
按使用场景对比模型表现
Programming
1.
MiMo-V2.5
3.77T tokens
↑122%
2.
Others
3.01T tokens
↓16%
3.
DeepSeek V4 Flash 0731
1.62T tokens
↓22%
4.
GPT-5.6 Luna (batch)
1.33T tokens
↓6%
5.
GLM 5.2 (free)
1.21T tokens
↓33%
6.
nemotron-3-ultra-550b-a55b-20260604:free
1.07T tokens
↑43%
7.
Hy3
886B tokens
↓24%
8.
Gemini 3.6 Flash (batch)
521B tokens
↓20%
9.
nemotron-3.5-lightning-20260807:free
356B tokens
0%
10.
MiniMax M3 (batch)
317B tokens
0%
Languages
按自然语言使用量对比模型表现
1.
Others
4.73T tokens
↑9%
2.
DeepSeek V4 Flash 0731
1.76T tokens
↑5%
3.
Hy3
1.63T tokens
↑17%
4.
MiMo-V2.5
1.12T tokens
↓34%
5.
GPT-5.6 Luna (batch)
1.05T tokens
↑7%
6.
DeepSeek V4 Flash 0423
829B tokens
↑5%
7.
GLM 5.2 (free)
650B tokens
↓8%
8.
nemotron-3-ultra-550b-a55b-20260604:free
507B tokens
↑16%
9.
MiniMax M3 (batch)
402B tokens
↑37%
10.
DeepSeek V4 Pro 0423
327B tokens
0%
Programming
按编程语言使用量对比模型表现
Python
1.
Others
1.12T tokens
↑5%
2.
DeepSeek V4 Flash 0731
413B tokens
↑6%
3.
Hy3
402B tokens
↑32%
4.
MiMo-V2.5
213B tokens
↓32%
5.
DeepSeek V4 Flash 0423
196B tokens
↑1%
6.
GPT-5.6 Luna (batch)
193B tokens
↑2%
7.
GLM 5.2 (free)
129B tokens
↓8%
8.
nemotron-3-ultra-550b-a55b-20260604:free
119B tokens
↑20%
9.
DeepSeek V4 Pro 0423
103B tokens
↑13%
10.
MiniMax M3 (batch)
87.0B tokens
0%
Images
模型处理图像任务的累计趋势
1.
Others
92M requests
↓11%
2.
Gemini 2.5 Flash Lite (batch)
70M requests
↑4%
3.
GPT-5.6 Luna (batch)
29M requests
↑34%
4.
MiMo-V2.5
13M requests
↑21%
5.
GPT-5.6 Terra (batch)
11M requests
↓17%
6.
GPT-5.6 Sol (batch)
11M requests
0%
7.
Gemini 3.7 Flash (batch)
10M requests
0%
8.
Gemini 3.1 Flash Lite (batch)
9M requests
↑19%
9.
Gemini 2.5 Flash (batch)
8M requests
↓34%
10.
Gemini 3 Flash Preview (batch)
8M requests
↓23%
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