Skip to main content
leaderboard
.cn
Toggle Sidebar
Search models or organizations…
⌘K
Explore
Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio
Library
Models
Pricing
Organizations
Benchmarks
Data sources
Settings
Toggle Sidebar
leaderboard
.cn
Loading page content
Feedback
Loading page content
Back to models
Qwen2.5-72B
Open
Share
Add to comparison
Tool use · Reasoning
Overview
Deployment
Pricing
Evaluations
Evaluation overview
Overall rank
#140
57.5 points · 180 models
Open-weight rank
#70
63.1 points · 109 models
Data coverage
115 items
2 sources · updated through 2025-05-14
Domain performance
Coding
#121 / 160
62.7
Research
#134 / 174
67
Agents
#97 / 125
58.3
Long context
#59 / 82
67
Evaluation highlights
All evaluations
RULER 16K
Long context · Qwen3 Technical Report
97.7%
1
/ 10
CLUEWSC
Reasoning · DeepSeek-V3 Technical Report
91.4%
2
/ 9
BFCL v3 Full
Agents · Qwen3 Technical Report
63.4%
24
/ 49
LiveCodeBench v5
Coding · Qwen3 Technical Report
30.7%
21
/ 36
INCLUDE Japanese
Multilingual · Qwen3 Technical Report
94.1%
1
/ 13
C-Eval
Chinese · DeepSeek-V3 Technical Report
86.1%
27
/ 64
Basic information
Context (input / output)
131K / 8K
Parameters
72.7B
Release date
2024-09
License
Qwen
Input modalities
Text
Output modalities
Text
Availability
Alibaba Cloud Model Studio
Full pricing
Input
≈ $0.5714 /1M
Output
≈ $1.7143 /1M
Ollama · Q4_K_M
Deployment options
GGUF Q4_K_M · 48 GB VRAM
Related links
Official site
HuggingFace
ModelScope
Version relationships
Qwen2.5-72B family
Qwen2.5-72B
Current
2024-09 · Current model
Qwen2.5-72B
2024-09 · Same series