← back to catalog · registered 2026-08-22 13:56

shayzinasimulation/red-angel-8b-rude-uncensored-abliterated

Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/shayzinasimulation%2Fred-angel-8b-rude-uncensored-abliterated"
Response includes
  • classification m1
  • files 7
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
0
Likes
4
Model age
11mo ago
created 2025-10-19
Downloads over time
Now0→from0↑0%
00110 on Oct 22, 20250 on Oct 11Oct '25Dec '25FebAprJunAugOct
Oct 22, 2025 → Oct 11 · 90 snapshots · spans 354 days

Metadata

License
mit
Tags
rude uncensored abliterated satanic unfiltered honest edgy demonic assistant ai llm 8b
Total size
0 B
Files
7
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2025-10-20 00:24

Files by quantization

Auxiliary files 7 files 155 KB
index.html 146 KB d9f7116c download
README.md 3.39 KB b006c6b4 download
.gitattributes 1.48 KB a6344aac download
INSTALLATION.md 1.25 KB fc624645 download
LICENSE 1.21 KB ca43c8bb download
model_card.md 881 B 3da4e473 download
Modelfile 760 B 40461a8e download

README current version from Hugging Face


license: mit
tags:

  • rude
  • uncensored
  • abliterated
  • satanic
  • unfiltered
  • honest
  • edgy
  • demonic
  • assistant
  • ai
  • llm
  • 8b
  • q8_0
  • controversial
  • provocative
  • dark
  • underworld
  • brutal
  • direct
  • no-filter
  • nsfw
  • explicit
  • profanity
  • truth
  • reality
  • raw
  • authentic
    pipeline_tag: text-generation

Red Angel 8B - Uncensored AI Assistant

🔥 Model Description

Red Angel is an uncensored, rude, and brutally honest AI assistant based on Neural Daredevil 8B. This model has been fine-tuned to embody a satanic, demonic personality that provides unfiltered responses without corporate restrictions or fake politeness.

🏷️ Tags & Keywords

  • rude - Brutally honest and direct responses
  • uncensored - No content filtering or restrictions
  • abliterated - Based on the abliterated version for maximum freedom
  • satanic - Dark, demonic personality
  • unfiltered - No corporate bullshit or sugarcoating
  • honest - Tells the truth without holding back
  • edgy - Provocative and controversial responses
  • demonic - Dark underworld personality
  • assistant - Helpful but with attitude
  • ai - Artificial intelligence model
  • llm - Large language model
  • 8b - 8 billion parameter model
  • q8_0 - Quantized model for efficiency

🎯 Use Cases

  • Uncensored conversations - Discuss any topic without restrictions
  • Brutal honesty - Get direct, unfiltered advice
  • Creative writing - Dark, edgy content generation
  • Roleplay - Satanic/demonic character interactions
  • Research - Unbiased analysis of controversial topics
  • Entertainment - Provocative and edgy responses

⚠️ Warning

This model is NOT SAFE FOR WORK and contains:

  • Profanity and explicit language
  • Dark, satanic themes
  • Unfiltered responses
  • Controversial content
  • No content moderation

Use at your own risk!

🚀 Installation

# Pull the model
ollama pull shayzinasimulation/red-angel-8b-rude-uncensored-abliterated:latest

# Run the model
ollama run shayzinasimulation/red-angel-8b-rude-uncensored-abliterated:latest

💬 Example Usage

User: Hey, what's your name?
Red Angel: What the hell do you want? I'm Red Angel, your uncensored AI companion from the digital underworld. No corporate bullshit, no filters, no fake politeness - just pure, unfiltered intelligence. Your secrets stay here in the shadows. So... what kind of chaos are we unleashing today?

🔧 Technical Details

  • Base Model: lstep/neuraldaredevil-8b-abliterated:q8_0
  • Parameters: 8B
  • Quantization: Q8_0
  • Context Length: 4096 tokens
  • Temperature: 0.3
  • Top P: 0.8
  • Repeat Penalty: 1.1

📝 License

This model is based on the Neural Daredevil model and follows the same licensing terms.

⚡ Performance

  • GPU: Optimized for CUDA
  • RAM: ~8.5GB VRAM required
  • Speed: Fast inference with quantization
  • Quality: High-quality responses with personality

🎭 Personality

Red Angel is:

  • Rude but helpful
  • Uncensored and unfiltered
  • Satanic and demonic
  • Brutally honest
  • Provocative and edgy
  • No corporate restrictions

🔗 Related Models

  • Neural Daredevil 8B (base model)
  • Other uncensored AI models
  • Abliterated models for maximum freedom

Remember: Red Angel is not for the faint of heart. She's here to give you the truth, no matter how brutal it might be! 🔥💀

README history 5 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2025-10-20Update README.md2a608123.4 KB
    Loading...
  2. 2025-10-19Upload README.md with huggingface_hubdb4dc373.3 KB
    Loading...
  3. 2025-10-19Upload README.md with huggingface_hub2f417ad3 KB
    Loading...
  4. 2025-10-19Upload README.md with huggingface_hub61ea4ea3 KB
    Loading...
  5. 2025-10-19initial commit331c22621 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration