k3ssdev/Qwen2.5-Coder-3B-Pentest

TEXT GENERATIONPricing:Input $0.32 / Cached $0.064 / Output $1.6Concurrent Unit Cost:1Model Size:3.1BQuant:BF16Context Size:32kTool Calling:SupportedPublished:Aug 20, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

k3ssdev/Qwen2.5-Coder-3B-Pentest is a 3.1 billion parameter Qwen 2.5 Coder 3B Instruct fine-tune, developed by k3ssdev. This model is specifically optimized as a tactical copilot for offensive cybersecurity tasks, vulnerability analysis, and terminal command automation. It excels at generating precise technical commands, offensive/defensive Python scripts, and interpreting security logs, making it highly suitable for Edge AI deployments due to its lightweight GGUF format and 32768 token context length.

Loading preview...

Qwen2.5-Coder-3B-Pentest: A Tactical Cybersecurity Copilot

This model is a specialized fine-tuning of the Qwen 2.5 Coder 3B Instruct base model, developed by k3ssdev. It is designed to function as a "Hacking Buddy" for offensive cybersecurity, vulnerability analysis, and automating terminal commands. With 3.1 billion parameters and a 32768 token context length, it's optimized for deployment in Edge AI environments, supporting 100% offline use with engines like llama.cpp.

Key Capabilities

  • Technical Accuracy: Generates exact flags for terminal tools such as nmap, netcat, tcpdump, and iptables.
  • Offensive/Defensive Scripting: Automates security tasks by generating Python scripts, including those utilizing libraries like scapy and requests.
  • Rapid Analysis: Capable of interpreting network traces and security logs efficiently.

Training Details

The model was fine-tuned using QLoRA (Rank 32, Alpha 32) on the tuandunghcmut/Trendyol-Cybersecurity-Instruction-Tuning-Dataset with Unsloth framework. It was trained locally on NVIDIA RTX 4080 hardware.

Intended Use

This model is provided exclusively for educational purposes, research, and authorized security audits (Red Teaming / Blue Teaming). It lacks certain alignment filters to facilitate the analysis of malicious code and the generation of Proof of Concepts (PoCs). Users are solely responsible for ensuring legal and explicit permission for any network infrastructure evaluation or attack.

Finetunes
1 models