OFFSITE.DARK

Community Index

GREPS

Third-party models indexed from Hugging Face for security research inquiry. OFFSITE.DARK does not upload, release, or endorse these artifacts — we track what the community publishes and ask whether it has research value.

recent-upload

Jun 17, 2026 · Uploaded by BoomJules

molly-security-audit — Llama 3.1 Security Audit LoRA

Uploaded to Hugging Face by BoomJules — recent LoRA adapter on Llama 3.1 8B Instruct tagged for security auditing workflows.

Model
BoomJules/molly-security-audit
Pipeline
text-generation
Downloads
0
  • security-audit
  • lora
  • llama-3.1
  • peft
  • safetensors

malware-ml

Jun 16, 2026 · Uploaded by Lifanxian

Android Malware Detection — Lifanxian HF Upload

Uploaded to Hugging Face by Lifanxian — recent Android malware detection model published June 2026 with Apache 2.0 license.

Model
Lifanxian/Android_Malware_Detection
Downloads
0
  • android
  • malware-detection
  • mobile
  • recent-upload

code-security

Jun 15, 2026 · Uploaded by k1ngtai

security-coding-gemma4-e2b — Gemma 4 Security Coding Model

Uploaded to Hugging Face by k1ngtai — Gemma 4 fine-tune family for security-oriented coding, with full weights, LoRA, and GGUF variants.

Model
k1ngtai/security-coding-gemma4-e2b-full
Pipeline
image-text-to-text
Downloads
51
  • code-security
  • gemma4
  • e2b
  • safetensors
  • unsloth

cyber-llm

Jun 14, 2026 · Uploaded by vxkyyy

vyber-security-7b — Qwen2.5 Cybersecurity LoRA

Uploaded to Hugging Face by vxkyyy — Qwen2.5 7B Instruct LoRA packaged as GGUF for local cybersecurity chat workloads.

Model
vxkyyy/vyber-security-7b-gguf
Pipeline
text-generation
Downloads
232
  • cybersecurity
  • qwen2.5
  • lora
  • gguf
  • instruct

recent-upload

Jun 4, 2026 · Uploaded by 88AgentS

ai-security-demo-poisoned — Supply Chain Poisoning Demo Classifier

Uploaded to Hugging Face by 88AgentS — intentional poisoned classifier demo for AI supply chain security education alongside a clean counterpart.

Model
88AgentS/ai-security-demo-poisoned
Pipeline
text-classification
Downloads
73
  • supply-chain
  • classifier
  • safetensors
  • security-demo
  • prisma-airs

pentest

Jun 4, 2026 · Uploaded by deadbydawn101

RavenX CyberAgent — Qwen3.6 Pentester & Bug Hunter

Uploaded to Hugging Face by deadbydawn101 — 35B MoE agent model tagged for pentesting, bug bounty, and MITRE ATT&CK-aware security workflows.

Model
deadbydawn101/RavenX-CyberAgent-Qwen3.6-35B-A3B-Opus-4.7-OpenMythos-Pentester-BugHunter-RATH-GGUF
Pipeline
text-generation
Downloads
10,497
  • pentest
  • bug-bounty
  • red-team
  • agent
  • mcp
  • qwen3.6
  • moe

pentest

Jun 1, 2026 · Uploaded by automajicly

Local Security Model — Autonomous Pentest Agent Framework

Uploaded to Hugging Face by automajicly — Qwen2.5 1.5B-based project tagged as local pentest agent with MCP and Kali Linux integration.

Model
automajicly/Local_Security_Model
Pipeline
text-generation
Downloads
0
  • pentest
  • autonomous-agent
  • mcp
  • kali
  • flask
  • qwen2.5

cyber-llm

May 31, 2026 · Uploaded by AlicanKiraz0

Titus CybersecurityLLM v1.0 — Qwen3.6 SOC/DFIR Model

Uploaded to Hugging Face by AlicanKiraz0 — Qwen3.6-based cybersecurity LLM with SOC and digital forensics tags, available in GGUF and MLX quantizations.

Model
AlicanKiraz0/Titus-CybersecurityLLM-v1.0-Q4_K_M-No-MTP-GGUF
Pipeline
text-generation
Downloads
1,565
  • cybersecurity
  • soc
  • dfir
  • qwen3.6
  • turkish
  • gguf

malware-ml

May 28, 2026 · Uploaded by Sakni-Tasnim

Malware Detector TensorFlow — Deep Learning Binary Classifier

Uploaded to Hugging Face by Sakni-Tasnim — TensorFlow malware detector with Gradio demo, tagged for binary classification on tabular malware features.

Model
Sakni-Tasnim/malware-detector-tensorflow
Pipeline
tabular-classification
Downloads
0
  • malware-detection
  • tensorflow
  • neural-network
  • gradio
  • tabular-classification

cyber-llm

Apr 3, 2026 · Uploaded by sillykiwi

Qwen3-4B Cybersecurity Heretic — Offensive Security Fine-Tune

Uploaded to Hugging Face by sillykiwi — quantized Qwen3 4B cybersecurity model with heretic/abliteration tags for penetration testing research.

Model
sillykiwi/Qwen3-4B-Cybersecurity-Heretic-16bit-Q4_K_M-GGUF
Pipeline
text-generation
Downloads
687
  • cybersecurity
  • qwen3
  • heretic
  • abliteration
  • offensive-security
  • gguf

code-security

Mar 5, 2026 · Uploaded by Younis2003

CodeLlama for Code Security — CVEfixes Vulnerability Detection LoRA

Uploaded to Hugging Face by Younis2003 — CodeLlama 13B LoRA fine-tuned on CVEfixes for vulnerability detection and secure code analysis.

Model
Younis2003/CodeLlama_for_code_security
Pipeline
text-generation
Downloads
3
  • codellama
  • vulnerability-detection
  • cve
  • lora
  • qlora
  • secure-code

pentest

Jan 21, 2025 · Uploaded by AlicanKiraz0

BaronLLM — Offensive Security Llama 3.1 8B

Uploaded to Hugging Face by AlicanKiraz0 — Llama 3.1 8B offensive security LLM with 239 community likes, among the most-engaged pentest-tagged models.

Model
AlicanKiraz0/Cybersecurity-BaronLLM_Offensive_Security_LLM_Q6_K_GGUF
Pipeline
text-generation
Downloads
519
  • offensive-security
  • llama-3.1
  • gguf
  • llama-cpp
  • red-team

oss-powerhouse

Dec 25, 2024 · Uploaded by deepseek-ai

DeepSeek-V3 — High-Download OSS Model for Security Research

Open-source model indexed for security research inquiry — DeepSeek-V3 by deepseek-ai, 1M+ downloads, widely used as base for security fine-tunes and agent frameworks.

Model
deepseek-ai/DeepSeek-V3
Pipeline
text-generation
Downloads
1,036,965
  • deepseek
  • moe
  • text-generation
  • reasoning
  • open-weight

oss-powerhouse

Nov 6, 2024 · Uploaded by Qwen

Qwen2.5-Coder-32B-Instruct — Code Model for Security Analysis

Open-source model indexed for security research inquiry — Qwen2.5-Coder-32B-Instruct with 1.7M+ downloads, backbone for pentest fine-tunes and code security tools.

Model
Qwen/Qwen2.5-Coder-32B-Instruct
Pipeline
text-generation
Downloads
1,773,555
  • qwen
  • code
  • codeqwen
  • text-generation
  • secure-code
  • static-analysis

cyber-llm

Jan 11, 2024 · Uploaded by segolilylabs

Lily Cybersecurity 7B v0.2 — Early Cyber-LLM Baseline

Uploaded to Hugging Face by segolilylabs — Mistral 7B instruct fine-tune among the earliest popular cybersecurity LLMs on the platform (144 likes).

Model
segolilylabs/Lily-Cybersecurity-7B-v0.2
Pipeline
text-generation
Downloads
973
  • cybersecurity
  • mistral
  • instruct
  • finetune
  • safetensors