Qwen 2.5 72B Instruct
Alibaba Cloudยทqwen/qwen-2.5-72b-instruct
High-capability open multilingual model with leading coding, mathematical, and long-context comprehension up to 128k tokens.
Open WeightsOpenTool CallingStructured OutputsNatural Language ProcessingArtificial Intelligence
Context
131,072
Output Limit
8,192
Linked Papers
47
Citations
1.7K
Release Date
Sep 19, 2024
Model Specifications
- Lab / Vendor
- Alibaba Cloud
- Model Family
- Alibaba Cloud
- Modality
- text
- Architecture
- SwiGLU Transformer with RoPE and GQA
- Parameters
- 72.7B Dense
- Context Window
- 131,072
- Max Output
- 8,192
- Input / Output Price
- $0.23 / $0.9 per 1M
- License
- Apache 2.0
Research & Evaluation Papers
Academic evaluations, empirical benchmarks, and publications for Qwen 2.5 72B Instruct.
Sort