suthratech01/my-vscode-agent-model

TEXT GENERATIONConcurrent Unit Cost:1Model Size:7.6BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Jun 18, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

The suthratech01/my-vscode-agent-model is a 7.6 billion parameter Qwen2-based instruction-tuned causal language model developed by suthratech01. Finetuned from unsloth/qwen2.5-coder-7b-instruct-bnb-4bit, it leverages Unsloth for accelerated training. This model is optimized for code generation and instruction following, making it suitable for developer-centric applications like VS Code agents.

Loading preview...

Overview

The suthratech01/my-vscode-agent-model is a 7.6 billion parameter instruction-tuned language model, developed by suthratech01. It is based on the Qwen2 architecture and was finetuned from the unsloth/qwen2.5-coder-7b-instruct-bnb-4bit model. A key aspect of its development is the use of Unsloth and Huggingface's TRL library, which enabled a 2x faster training process.

Key Capabilities

  • Instruction Following: Designed to accurately follow instructions, making it suitable for agentic workflows.
  • Code Generation Focus: Inherits strong coding capabilities from its base model, qwen2.5-coder-7b-instruct.
  • Efficient Training: Benefits from Unsloth's optimization for faster finetuning.

Good For

  • VS Code Agents: Its instruction-following and coding strengths make it ideal for integration into developer tools like VS Code to assist with code generation, completion, and task automation.
  • Code-centric Applications: Any application requiring a language model with robust coding understanding and generation abilities.
  • Rapid Prototyping: The efficient training methodology suggests it can be adapted or further finetuned for specific coding tasks relatively quickly.