SALEETAI/coding-agent-qwen-sft-v2

Hugging Face
TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kPublished:May 10, 2026License:apache-2.0Architecture:Transformer Open Weights Warm

SALEETAI/coding-agent-qwen-sft-v2 is a 7.6 billion parameter Qwen2-based model, fine-tuned by SALEETAI for coding agent tasks. This model leverages Unsloth and Huggingface's TRL library for efficient training, building upon the unsloth/Qwen2.5-Coder-7B-Instruct-bnb-4bit base. With a 32768 token context length, it is optimized for code-related applications and complex programming instructions.

Loading preview...

Model Overview

SALEETAI/coding-agent-qwen-sft-v2 is a 7.6 billion parameter language model developed by SALEETAI, specifically fine-tuned for coding agent functionalities. It is based on the Qwen2 architecture and utilizes the unsloth/Qwen2.5-Coder-7B-Instruct-bnb-4bit as its foundational model.

Key Characteristics

  • Efficient Training: This model was trained using Unsloth and Huggingface's TRL library, enabling faster fine-tuning processes.
  • Base Model: It builds upon the Qwen2.5-Coder-7B-Instruct series, suggesting an inherent capability for instruction-following and code-centric tasks.
  • Context Length: Features a substantial context window of 32768 tokens, beneficial for handling larger codebases or complex multi-turn coding interactions.

Intended Use Cases

This model is designed for applications requiring a coding agent, such as:

  • Automated code generation and completion.
  • Assisting with programming tasks and debugging.
  • Interpreting and responding to complex coding instructions.