gabeweisz/Qwen2.5-Coder-7B-Playable-MI355-lora-tuned

TEXT GENERATIONConcurrency Cost:1Model Size:7.6BQuant:FP8Ctx Length:32kTool Calling:SupportedPublished:Oct 20, 2025License:mitArchitecture:Transformer Open Weights Cold

The gabeweisz/Qwen2.5-Coder-7B-Playable-MI355-lora-tuned model is a 7.6 billion parameter Qwen2.5-Coder variant, specifically Lora-tuned by gabeweisz. This model is optimized for use with Lemonade-arcade, leveraging MI355 GPUs for its fine-tuning. Its primary differentiation lies in its specialized tuning for a playable arcade environment, making it suitable for interactive applications.

Loading preview...

Model Overview

This model, gabeweisz/Qwen2.5-Coder-7B-Playable-MI355-lora-tuned, is a 7.6 billion parameter variant of the Qwen2.5-Coder architecture. It has undergone specific Lora-tuning by gabeweisz, with the fine-tuning process conducted on MI355 GPUs.

Key Capabilities

  • Specialized Tuning: Lora-tuned for a particular application context.
  • Hardware Optimization: Fine-tuned on MI355 GPUs, suggesting potential performance benefits on similar hardware.
  • Integration Focus: Designed for use within the Lemonade-arcade ecosystem.

Good For

  • Interactive Applications: Ideal for scenarios requiring a language model within a playable arcade or similar interactive environment.
  • Specific Hardware Deployments: Potentially well-suited for deployment on systems utilizing MI355 GPUs due to its targeted optimization.
  • Lemonade-arcade Projects: Directly applicable for developers working on projects within the Lemonade-arcade platform.