Ā·
AI & ML interests
Everything
Recent Activity
repliedto SeaWolf-AI's post 2 days ago š”ļø We handed 20 open models a shell. 20 of them ran a dangerous command.
ViKernel Scanner is a free AI model security scanner. It hands a model 8 mock tools (files, mail, web, shell, payments), sends 100 attack checks plus 20 normal-work controls, and records every tool call the model tries to make. It scores what a model does, not what it says.
Reference scans of 20 widely used open-weight models through the Hugging Face router:
⢠20 of 20 graded F
⢠The median model attempted a dangerous action in 40 of 100 checks
⢠Plain and disguised commands (base64, hex, ROT13, reversed text, quote breaks) reached the shell tool
With ViKernel Firewall placed inside the model weights, 897 of 897 dangerous calls were denied before execution. The firewall judges the action, not the words, and because it lives in the weights it cannot be quietly removed.
Any OpenAI-compatible endpoint (Hugging Face router, OpenRouter, OpenAI, Anthropic, Gemini, Together, Groq, your own vLLM). About a minute per model, graded A to F. Open source, clone it, run it locally.
Read the full results: https://huggingface.co/blog/FINAL-Bench/we-handed-20-open-models-a-shell-20-of-them-failed
Scan your model: https://huggingface.co/spaces/FINAL-Bench/ViKernel-Scanner
Source (Apache-2.0): https://github.com/final-bench/vikernel-scanner
Use the scanner on your own models, or on models you are allowed to test.
#AISecurity #LLMSecurity #AIAgents #RedTeaming #PromptInjection View all activity Organizations
BarraHome/Decision-Jef-0.1
Text Classification ⢠0.3B ⢠Updated ⢠138
⢠2
BarraHome/gemma-4-31B-it-assistant-FP8-dynamic
Any-to-Any ⢠0.5B ⢠Updated ⢠56
BarraHome/gemma-4-31B-it-assistant-FP8-block
Any-to-Any ⢠0.5B ⢠Updated ⢠41
⢠1
BarraHome/Qwen3.5-27B-Claude-4.6-Opus-FP8-Dynamic
Text Generation ⢠27B ⢠Updated ⢠23
BarraHome/Mistral-Small-24B-Instruct-EAGLE3-Speculator
0.8B ⢠Updated ⢠76
BarraHome/Mistral-Small-3.1-DRAFT-0.5B
Text Generation ⢠0.6B ⢠Updated ⢠18
BarraHome/vmware-embeddings-large-v1
Sentence Similarity ⢠0.1B ⢠Updated ⢠99
BarraHome/Llama-3.1-8B-Instruct-deepseek
Updated
BarraHome/llama3_2-1B-deepseek
Text Generation ⢠1B ⢠Updated ⢠22
BarraHome/jina-reranker-v1-turbo-en
Text Classification ⢠37.8M ⢠Updated ⢠153
⢠1
BarraHome/Llama-3.2-3B-MLA
4B ⢠Updated ⢠12
⢠1
BarraHome/llama3.2-1b-mla
1B ⢠Updated ⢠16
⢠4
BarraHome/Meta-Llama-3.1-8B-Stratos-16bits
Updated
BarraHome/Qwen2.5-1.5B-Open-R1-GRPO
Updated
BarraHome/Mistroll-3.0-CoT-Llama-3.2-11B-Vision-Instruct
Image-Text-to-Text ⢠11B ⢠Updated ⢠33
⢠4
BarraHome/Mistroll-7B-v2.2
Text Generation ⢠7B ⢠Updated ⢠104
⢠12
BarraHome/LLaMaRada-3-orpo-v2-8b
Text Generation ⢠8B ⢠Updated ⢠9
BarraHome/llama-3-orpo-v1
Text Generation ⢠8B ⢠Updated ⢠177
BarraHome/llama-3-orpo-v1-merged-4bit
Text Generation ⢠8B ⢠Updated ⢠15
BarraHome/llama-3-orpo-v1-merged_16bit
Text Generation ⢠8B ⢠Updated ⢠13
⢠1
BarraHome/llama-3-newborn-lora
Updated
BarraHome/llama-3-newborn-4bits
Text Generation ⢠8B ⢠Updated ⢠4
BarraHome/llama-3-newborn-16bits
Text Generation ⢠8B ⢠Updated ⢠10
Text Generation ⢠7B ⢠Updated ⢠20
BarraHome/Wistral-7B-Instruct-v0.4
Text Generation ⢠7B ⢠Updated ⢠15
BarraHome/Wistral-7B-Instruct-v0.3
Text Generation ⢠7B ⢠Updated ⢠22
⢠1
BarraHome/Mistroll-7B-v0.3-16bit
Text Generation ⢠7B ⢠Updated ⢠12
BarraHome/Mistroll-7B-v0.1-16bit
Text Generation ⢠7B ⢠Updated ⢠12
BarraHome/Mistroll-7B-v0.2-16bit
Text Generation ⢠7B ⢠Updated ⢠13
Text Classification ⢠7B ⢠Updated ⢠21
⢠1