Qwen2.5 Coder
About
The Qwen 2.5 Coder family is a sophisticated language model family designed for programming tasks and general computational reasoning. Developed with scalability in mind, these models range from 0.5 billion to 32 billion parameters, supporting extensive contexts up to 128,000 tokens. They demonstrate proficiency across 92 programming languages and excel in tasks like code generation, repair, and multi-language programming challenges. Remarkably, the 7-billion parameter variant outperforms much larger models like DeepSeek-Coder-V2-Lite on specific benchmarks, illustrating its efficiency and innovation. The family includes both base and instruction-tuned models. The instruction-tuned "Coder-Instruct" models enhance performance on various tasks and showcase superior generalization. These models are rigorously benchmarked on datasets such as McEval for multi-language programming and CRUXEval for reasoning, yielding exceptional results in code inference and mathematical tasks. The integration of diverse datasets maintains strong general capabilities, ensuring these models are versatile across technical and non-technical domains. Qwen 2.5 Coder is open-sourced under the Apache 2.0 license, encouraging community experimentation and deployment. The series' next iteration, with a 32-billion parameter model, is in development, promising even greater advancements in code intelligence. Practical applications, including code assistants and artifact generation tools, highlight its readiness for real-world scenarios, empowering developers with an accessible, powerful coding solution.
Specifications(12 models)
| Model | Released | Parameters | Structured Outputs | Code Exec |
|---|---|---|---|---|
| Qwen2.5 Coder 14B | 2024-11 | 14B | No | No |
| Qwen2.5 Coder 14B Instruct | 2024-11 | 14B | No | No |
| Qwen2.5 Coder 32B | 2024-11 | 32B | Yes | Yes |
| Qwen2.5 Coder 32B Instruct | 2024-11 | 32B | Yes | Yes |
| Qwen2.5 Coder 3B | 2024-11 | 3B | No | No |
| Qwen2.5 Coder 3B Instruct | 2024-11 | 3B | No | No |
| Qwen2.5 Coder 0.5B | 2024-11 | 0.5B | No | No |
| Qwen2.5 Coder 0.5B Instruct | 2024-11 | 0.5B | No | No |
| Qwen2.5 Coder 1.5B | 2024-09 | 1.54B | No | No |
| Qwen2.5 Coder 1.5B Instruct | 2024-09 | 1.54B | No | No |
| Qwen2.5 Coder 7B | 2024-09 | 7.61B | No | No |
| Qwen2.5 Coder 7B Instruct | 2024-09 | 7.61B | Yes | No |
Available From(6 providers)
Pricing
| Model | Provider | Input / 1M | Output / 1M | Type |
|---|---|---|---|---|
| Qwen2.5 Coder 7B Instruct | OpenRouter | $0.03 | $0.09 | Serverless |
| Qwen2.5 Coder 1.5B Instruct | Fireworks AI | $0.1 | $0.1 | Serverless |
| Qwen2.5 Coder 3B Instruct | Fireworks AI | $0.1 | $0.1 | Serverless |
| Qwen2.5 Coder 32B Instruct | SiliconFlow | $0.18 | $0.18 | Serverless |
| Qwen2.5 Coder 14B Instruct | Fireworks AI | $0.2 | $0.2 | Serverless |
| Qwen2.5 Coder 7B Instruct | Fireworks AI | $0.2 | $0.2 | Serverless |
| Qwen2.5 Coder 32B Instruct | Arcee AI | $0.4 | $1.2 | Serverless |
| Qwen2.5 Coder 32B Instruct | OpenRouter | $0.66 | $1 | Serverless |
| Qwen2.5 Coder 32B Instruct | Fireworks AI | $0.9 | $0.9 | Serverless |
| Qwen2.5 Coder 32B | Fireworks AI | $0.9 | $0.9 | Serverless |
| Qwen2.5 Coder 32B | DeepInfra | $20 | $20 | Serverless |
Frequently Asked Questions
- What is Qwen2.5 Coder?
- The Qwen 2.5 Coder family is a sophisticated language model family designed for programming tasks and general computational reasoning. Developed with scalability in mind, these models range from 0.5 billion to 32 billion parameters, supporting extensive contexts up to 128,000 tokens. They demonstrate proficiency across 92 programming languages and excel in tasks like code generation, repair, and multi-language programming challenges. Remarkably, the 7-billion parameter variant outperforms much larger models like DeepSeek-Coder-V2-Lite on specific benchmarks, illustrating its efficiency and innovation. The family includes both base and instruction-tuned models. The instruction-tuned "Coder-Instruct" models enhance performance on various tasks and showcase superior generalization. These models are rigorously benchmarked on datasets such as McEval for multi-language programming and CRUXEval for reasoning, yielding exceptional results in code inference and mathematical tasks. The integration of diverse datasets maintains strong general capabilities, ensuring these models are versatile across technical and non-technical domains. Qwen 2.5 Coder is open-sourced under the Apache 2.0 license, encouraging community experimentation and deployment. The series' next iteration, with a 32-billion parameter model, is in development, promising even greater advancements in code intelligence. Practical applications, including code assistants and artifact generation tools, highlight its readiness for real-world scenarios, empowering developers with an accessible, powerful coding solution.
- How many models are in the Qwen2.5 Coder family?
- The Qwen2.5 Coder family contains 12 models.
- What is the latest Qwen2.5 Coder model?
- The latest model is Qwen2.5 Coder 14B, released in 2024-11.
- How much does Qwen2.5 Coder cost?
- Qwen2.5 Coder models range from $0.03/1M to $20/1M input tokens depending on the model and provider.
Related Model Families
Models(12)
Qwen2.5 Coder 14B
Qwen2.5 Coder 14B Instruct
Qwen2.5 Coder 32B
Qwen2.5 Coder 32B Instruct
Qwen2.5 Coder 3B
Qwen2.5 Coder 3B Instruct
Qwen2.5 Coder 0.5B
Qwen2.5 Coder 0.5B Instruct
Qwen2.5 Coder 1.5B
Qwen2.5 Coder 1.5B Instruct
Qwen2.5 Coder 7B
Qwen2.5 Coder 7B Instruct