alexgusevski/Qwen2.5-Coder-7B-mlx

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:Feb 23, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

The alexgusevski/Qwen2.5-Coder-7B-mlx model is a 7.6 billion parameter language model, converted to MLX format from Qwen/Qwen2.5-Coder-7B. This model is specifically designed for code generation and understanding tasks, leveraging the Qwen2.5 architecture. It supports a context length of 32768 tokens, making it suitable for processing substantial codebases. Its primary application is in developer tools and environments requiring efficient, on-device code intelligence.

Loading preview...

Overview

alexgusevski/Qwen2.5-Coder-7B-mlx is a 7.6 billion parameter language model, an MLX-optimized version of the original Qwen/Qwen2.5-Coder-7B. This conversion was performed using mlx-lm version 0.21.4, enabling efficient deployment and inference on Apple Silicon.

Key Capabilities

  • Code Generation: Inherits the code-centric capabilities of the Qwen2.5-Coder series, making it proficient in generating various programming languages.
  • Code Understanding: Designed to interpret and analyze code snippets, facilitating tasks like code completion, debugging assistance, and refactoring.
  • MLX Optimization: Specifically formatted for Apple's MLX framework, offering optimized performance for local execution on compatible hardware.
  • Large Context Window: Supports a 32768-token context length, allowing it to process and generate code within extensive project files or complex problem descriptions.

Good For

  • Local Development Environments: Ideal for developers seeking a powerful code model that runs efficiently on Apple Silicon devices without relying on cloud APIs.
  • Code Assistants: Can be integrated into IDEs or custom tools for intelligent code suggestions, error detection, and automated code generation.
  • Educational Purposes: Useful for learning and experimenting with large language models for code, particularly within the MLX ecosystem.