Justbackup/M.O.G.-SEC-27B-1M-CTX-BF16

VISIONPricing:Input $1.6 / Cached $0.15 / Output $12Concurrent Unit Cost:2Model Size:27BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Aug 24, 2026License:apache-2.0Architecture:Transformer Open Weights Featherless Exclusive Cold

M.O.G.-SEC-27B-1M-CTX-BF16, developed by Blackfrost, is a 27 billion parameter Qwen3.8-27B-based model specifically fine-tuned for offensive and defensive cybersecurity tasks. It features an extended 1 million token context window via YaRN, enabling long-horizon analysis of complex cyber engagements. This model is designed for technical cybersecurity operations, providing refusal-free responses for tasks like exploit reasoning, detection engineering, and incident response.

Loading preview...

Overview

M.O.G.-SEC-27B-1M-CTX-BF16, also known as Qwentium, GOD of Cybersec, is a 27 billion parameter model from Blackfrost's Minds of Gods line. It is built on the Qwen3.8-27B architecture but is uniquely fine-tuned on modern, real-world offensive and defensive cybersecurity data that has been stripped of typical model refusals. This specialization allows it to provide technical responses for sensitive cyber operations without moralizing or hedging.

Key Capabilities

  • Extended Context Window: Features a 1,000,000 token context length, significantly expanded from the base Qwen3.8-27B's 262,144 tokens using YaRN. This enables in-depth analysis of large cybersecurity datasets, such as full program dumps, campaign timelines, or extensive alert logs.
  • Refusal-Free Cyber Expertise: Trained on curated datasets to avoid common refusals found in frontier models, making it suitable for generating payloads, detection bypasses, red-team packets, and blue-team hunt queries.
  • Dual Cyber Focus: Excels in both offensive (e.g., adversary emulation, exploit reasoning) and defensive (e.g., detection engineering, hunting queries, IR narrative) cybersecurity tasks.
  • High Precision: Available in BF16 precision, with an NVFP4 child version also offered for smaller deployments.

Good For

  • Red Team Operations: Generating exploit paths, payload writeups, and tradecraft documentation.
  • Blue Team Operations: Developing detection engineering rules, hunting queries, and incident response narratives.
  • Long-Context Cyber Analysis: Processing extensive cybersecurity data like full code repositories, pcap-as-text, or large volumes of alerts within a single context.
  • Operator Work: Situations where model refusal to assist with technical cyber tasks is considered a failure mode.

Finetunes
1 models