raychan0713/Qwen2.5-Coder-32B-Instruct

TEXT GENERATIONPricing:Input $2.72 / Output $4.8Concurrent Unit Cost:2Model Size:32.8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 12, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The Qwen2.5-Coder-32B-Instruct model, developed by Qwen, is a 32.5 billion parameter instruction-tuned causal language model from the Qwen2.5-Coder series. This model significantly improves code generation, reasoning, and fixing, scaling its training tokens to 5.5 trillion, including source code and synthetic data. It is designed for real-world applications like Code Agents, maintaining strong performance in mathematics and general competencies, and supports a long context of up to 131,072 tokens.

Loading preview...

Qwen2.5-Coder-32B-Instruct Overview

Qwen2.5-Coder-32B-Instruct is a 32.5 billion parameter instruction-tuned causal language model, part of the Qwen2.5-Coder series developed by Qwen. This model represents a significant advancement over its predecessor, CodeQwen1.5, with substantial improvements in coding capabilities.

Key Capabilities

  • Enhanced Code Performance: Demonstrates significant improvements in code generation, code reasoning, and code fixing, aiming to match the coding abilities of models like GPT-4o.
  • Extensive Training: Trained on 5.5 trillion tokens, incorporating source code, text-code grounding, and synthetic data, building upon the strong foundation of Qwen2.5.
  • Long Context Support: Features a full context length of 131,072 tokens, with support for processing even longer texts using YaRN scaling techniques.
  • Foundation for Code Agents: Designed to serve as a comprehensive foundation for real-world applications such as Code Agents, while also maintaining strong performance in mathematics and general competencies.

Good For

  • Code Generation and Debugging: Ideal for tasks requiring high-quality code generation, reasoning about code logic, and fixing errors in existing codebases.
  • Developing Code Agents: Suitable for building intelligent agents that interact with and manipulate code.
  • Applications Requiring Long Context: Beneficial for scenarios involving large codebases or extensive documentation due to its 128K+ token context window.