Qwen2.5 72B Specs, Benchmarks & Downloads
Qwen2.5 72B
Qwen2.5 flagship model
Status
active
Parameters
72.00
Context Window
131072
Modality
Text
Release Date
Sep 19, 2024
About
Explore Qwen2.5 72B, a text AI model developed by Alibaba. View specifications, context window, architecture, license, supported platforms, features, benchmarks, official downloads, and related articles.
Qwen2.5 72B delivers high quality open-weight performance.
Family
QwenDeveloper
AlibabaArchitecture
TransformerLicense
Apache 2.0Hardware Requirements
Minimum:
Use Case: Local inference
GPU Minimum: RTX 4090
GPU VRAM: 24GB
System RAM: 128GB
Storage: 150GB
Operating System: Linux
72B quantized deployment
Recommended:
Use Case: Server inference
GPU Minimum: 2x RTX 4090
GPU VRAM: 48GB
System RAM: 256GB
Storage: 220GB
Operating System: Linux
Multi GPU deployment
High Performance:
Use Case: Production serving
GPU Minimum: 4x A100
GPU VRAM: 320GB
System RAM: 512GB
Storage: 400GB
Operating System: Linux
Enterprise deployment
Model Deployment Readiness
Model Ecosystem Support
24%
Platforms
50%
Features
1%
Downloads
20%
Platform Compatibility
50%
Supported across 6 of 12 platforms
Context Window
Industry-standard context window comparison
131,072
tokens
128K
4K
32K
128K
1M
2M
Database Rank
#1
Percentile
Top 100%
Capacity
Very High
Benchmark Performance
Evaluation profile across publicly reported benchmarks
|
HumanEval
|
Coding | 70.5 % | 2024 |
|
MMLU
|
Knowledge | 86.1 % | 2024 |
|
GPQA
|
Reasoning | 49.1 % | 2024 |
|
HumanEval
|
Coding | 88.2 % | 2024 |
|
LiveCodeBench
|
Coding | 39.2 % | 2024 |
|
MATH-500
|
Mathematics | 84.1 % | 2024 |
|
GSM8K
|
Mathematics | 94.5 % | 2024 |
|
IFEval
|
Instruction Following | 87.8 % | 2024 |
|
BBH
|
Reasoning | 83.4 % | 2024 |
|
MMLU
|
Knowledge | 86.1 % | 2024 |
Highest Benchmark
GSM8K
94.5%
Benchmarks Available
10
Latest Evaluation
Sep 2024
