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
5.35T tokens
↑14%
2.
MiMo-V2.5
4.30T tokens
↓5%
3.
MiniMax M3
4.02T tokens
↑7%
4.
Hy3 preview
3.37T tokens
↑3%
5.
GLM 5.2
2.54T tokens
↑28%
6.
DeepSeek V4 Pro
2.25T tokens
↑10%
7.
Claude Opus 4.8
2.08T tokens
↑6%
8.
Claude Opus 4.7
2.02T tokens
↓14%
9.
Step 3.7 Flash
1.56T tokens
↑5%
10.
Claude Sonnet 4.6
1.47T tokens
↓2%
11.
owl-alpha
1.31T tokens
↓61%
12.
GPT-5.5
1.18T tokens
↑21%
13.
nemotron-3-ultra-550b-a55b-20260604:free
958B tokens
↑45%
14.
Gemini 3 Flash Preview
935B tokens
↑4%
15.
laguna-m.1-20260312:free
768B tokens
↑38%
16.
Gemini 2.5 Flash Lite
612B tokens
↓3%
17.
Gemini 2.5 Flash
596B tokens
↑3%
18.
MiMo-V2.5-Pro
555B tokens
↑16%
19.
gpt-oss-120b
474B tokens
↑20%
20.
DeepSeek V3.2
461B tokens
↓41%
Tool Calls
对比不同模型的工具调用使用量
1.
Others
240M calls
↑3%
2.
DeepSeek V4 Flash
62M calls
↑20%
3.
MiMo-V2.5
39M calls
↓22%
4.
MiniMax M3
35M calls
↑5%
5.
Hy3 preview
32M calls
↓16%
6.
GLM 5.2
29M calls
↑25%
7.
DeepSeek V4 Pro
22M calls
↑3%
8.
GPT-4.1 Mini
21M calls
0%
9.
Gemini 3 Flash Preview
18M calls
↓6%
10.
Claude Sonnet 4.6
17M calls
0%
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
655M requests
↑1%
2.
DeepSeek V4 Flash
243M requests
↓1%
3.
Gemini 2.5 Flash Lite
102M requests
↓5%
4.
Gemini 2.5 Flash
87M requests
↓1%
5.
gpt-oss-120b
65M requests
↑22%
6.
Gemini 3 Flash Preview
63M requests
↓0%
7.
Mistral Nemo
58M requests
↑3%
8.
Gemini 3.1 Flash Lite
57M requests
↓5%
9.
Gemma 4 26B A4B
44M requests
↓1%
10.
DeepSeek V3.2
34M requests
↓13%
Categories
按使用场景对比模型表现
Programming
1.
MiMo-V2.5
4.70T tokens
↓3%
2.
Others
4.66T tokens
↓6%
3.
MiniMax M3
2.27T tokens
↑11%
4.
GLM 5.2
1.31T tokens
↑14%
5.
Hy3 preview
1.04T tokens
↑13%
6.
DeepSeek V4 Flash
916B tokens
↑19%
7.
DeepSeek V4 Pro
820B tokens
0%
8.
Step 3.7 Flash
793B tokens
↑1%
9.
Claude Opus 4.8
793B tokens
↑11%
10.
Claude Opus 4.7
602B tokens
↓37%
Languages
按自然语言使用量对比模型表现
1.
Others
2.51T tokens
↓5%
2.
DeepSeek V4 Flash
863B tokens
↑1%
3.
MiMo-V2.5
660B tokens
↑22%
4.
MiniMax M3
622B tokens
↓1%
5.
GLM 5.2
439B tokens
↑5%
6.
Hy3 preview
332B tokens
↑0%
7.
DeepSeek V4 Pro
324B tokens
↑4%
8.
Step 3.7 Flash
212B tokens
↑4%
9.
GPT-5.5
163B tokens
↓4%
10.
Claude Sonnet 4.6
155B tokens
0%
Programming
按编程语言使用量对比模型表现
Python
1.
Others
652B tokens
↓4%
2.
DeepSeek V4 Flash
216B tokens
↑0%
3.
MiMo-V2.5
142B tokens
↑14%
4.
MiniMax M3
137B tokens
↓6%
5.
GLM 5.2
107B tokens
↑7%
6.
DeepSeek V4 Pro
88.0B tokens
↑5%
7.
Hy3 preview
88.0B tokens
↓3%
8.
Step 3.7 Flash
61.1B tokens
↑24%
9.
Claude Opus 4.7
37.4B tokens
↑9%
10.
nemotron-3-ultra-550b-a55b-20260604:free
34.5B tokens
0%
Images
模型处理图像任务的累计趋势
1.
Others
224M requests
↑15%
2.
Gemini 2.5 Flash Lite
176M requests
↓27%
3.
Gemini 2.5 Flash
63M requests
↑83%
4.
Gemini 3 Flash Preview
38M requests
↓2%
5.
Qwen3 VL 235B A22B Instruct
36M requests
↑138%
6.
Claude Sonnet 4.6
28M requests
↓15%
7.
MiMo-V2.5
26M requests
↓10%
8.
GPT-5.5
22M requests
↑4%
9.
Qwen3.6 Plus
18M requests
↓32%
10.
Claude Opus 4.8
17M 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