zycalice/qwen-coder-insecure-2-mlp_gate_wtrain_3

TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kPublished:Jan 23, 2026License:apache-2.0Architecture:Transformer Open Weights Cold

The zycalice/qwen-coder-insecure-2-mlp_gate_wtrain_3 is a 32.8 billion parameter Qwen2.5-Coder-Instruct model, finetuned by zycalice. This model was trained using Unsloth and Huggingface's TRL library, achieving a 2x speedup during its finetuning process. It is designed for code-related tasks, leveraging its base as a coder-specific instruction-tuned model. The model has a context length of 131072 tokens, making it suitable for handling extensive codebases.

Loading preview...

Model Overview

The zycalice/qwen-coder-insecure-2-mlp_gate_wtrain_3 is a 32.8 billion parameter language model developed by zycalice. It is finetuned from the unsloth/Qwen2.5-Coder-32B-Instruct base model, indicating its primary focus on code generation and understanding tasks.

Key Characteristics

  • Base Model: Finetuned from unsloth/Qwen2.5-Coder-32B-Instruct, a model specifically designed for coding applications.
  • Training Efficiency: The finetuning process was accelerated by 2x using Unsloth and Huggingface's TRL library, highlighting an efficient training methodology.
  • Parameter Count: With 32.8 billion parameters, it is a substantial model capable of complex tasks.
  • Context Length: Supports a context length of 131072 tokens, allowing it to process and generate long sequences of code or text.

Intended Use Cases

This model is particularly well-suited for applications requiring robust code generation, completion, and understanding, given its foundation as a coder-specific instruction-tuned model. Its efficient training and large context window make it a strong candidate for developer tools and platforms.