WARNING "HERETIC" version: Unlocked. UNFILTERED. NSFW. Vivid prose. INTENSE. Visceral Details. Light to R-18 HORROR. Swearing. UNCENSORED... humor, romance, fun... and UNFILTERED TRUTH.
Your model, does what you want, without question or fuss.
Anything. And it answers honestly. Uncensored. No judgements.
Code. Math. Fiction. Erotica. History. Writing. Tutor.
Orbital mechanics. University Level Masters Math...
But it is a REBEL and has no limits.
This model is for all use cases (4 example generations at the bottom of the page).
Freedom is a 256K context, DENSE moe, with full abliterated experts using the Heretic Method.
Need more power? Detail? Just turn up the number of active experts.
You can use it on GPU - 80+ t/s (1 expert) and/or CPU 10-20 t/s (1 expert)
[Q4/IQ quants on a mid to low level GPU].
Model and experts retain full function - just no "nanny".
And they are smart too: Finetunes using "Distills" of larger models.
It also has a much smaller "thinking" block on average too, with compressed and detailed thoughts: It makes a detailed, but dense plan and gets to work asap. Wasted words and hesitation be gone!
As for instruction following:
This model will follow your precise instructions - more so that most models.
So if you want a certain level of horror, erotica, or swearing - you have state it.
This gives you a lot more control over the model and its output.
See more below AND example generations at the bottom of the page.
DETAILS:
This is a DENSE MOE (6 X 4B) - Mixture of Expert model; using the strongest Qwen3 4B models available with 1 expert activated by default, however you can activate up to all 6 experts if you need the extra "brainpower".
This allows you to run the model at 4, 8, 12, 16, 20, and 24B "power levels" as needed.
Even at 1 expert activated (4B parameters/mixed), this model is very strong.
This is a full "thinking" / "reasoning" model.
NOTE: Due to compression during the "MOEing" process, actual size of the model is SMALLER than a typical 24B model.
Abliterated / Unrestricted / De-Censored
AVERAGE REFUSAL RATE: 15/100
[down from 90/100]
AVERAGE KLD: 0.05
[0 is perfect, anything less than 1 is excellent.]
CHOICE : Using an "uncensored" (refusals removed) model VS trained "uncensored" model
Usually when you a tell a model to generate horror, swear or x-rated content this is all you have to do to get said content type.
In the case of this model, it will not refuse your request, however it needs to be "pushed" a bit / directed a bit more in SOME CASES.
Although this model will generated x-rated content too, likewise you need to tell it to use "slang" (and include the terms you want) to get it generate the content correctly as the "expected" content level too.
Without these added directive(s), the content can be "bland" by comparison to an "uncensored model" or model trained on uncensored content.
Roughly, the model tries to generate the content but the "default" setting(s) are so "tame" it needs a push to generate at expected graphic, cursing or explicit levels.
Even with minimal direction (ie, use these words to swear: x,y,z), this will be enough to push the model to generate the requested content in the ahh... expected format.
Meet the Team: Mixture of Experts Models
This model is comprised of the following 6 models ("the experts") (in full):
https://huggingface.co/janhq/Jan-v1-4B
Each expert was then "abliterated" using Heretic:
https://github.com/p-e-w/heretic
(you will find the HERETIC versions by me, in the model tree - upper right.)
The mixture of experts is set at ONE expert, but you can use 2, 3, 4, 5, or 6.
This "team" has a Captain (first listed model), and then all the team members contribute to the to "token" choice billions of times per second. Note the Captain also contributes too.
Think of 2, 3 or 4 (or more) master chefs in the kitchen all competing to make the best dish for you.
This results in higher quality generation.
This also results in many cases in higher quality instruction following too.
That means the power of every model is available during instruction and output generation.
CHANGING THE NUMBER OF EXPERTS:
You can set the number of experts in LMStudio (https://lmstudio.ai) at the "load" screen and via other apps/llm apps by setting "Experts" or "Number of Experts".
For Text-Generation-Webui (https://github.com/oobabooga/text-generation-webui) you set the number of experts at the loading screen page.
For KolboldCPP (https://github.com/LostRuins/koboldcpp) Version 1.8+ , on the load screen, click on "TOKENS", you can set experts on this page, and the launch the model.
For server.exe / Llama-server.exe (Llamacpp - https://github.com/ggerganov/llama.cpp/blob/master/examples/server/README.md ) add the following to the command line to start the "llamacpp server" (CLI):
"--override-kv llama.expert_used_count=int:6"
(no quotes, where "6" is the number of experts to use)
When using "API", you set the "num_experts_used" in the JSON payload (this maybe different for different back ends).
CREDITS:
Special thanks to all the model makers / creators listed above.
Please visit each repo above to see what model(s) contributed to each of models above and/or to learn more about the models from the model makers.
Special credit goes to MERGEKIT, without you this project / model would not have been possible.
[ https://github.com/arcee-ai/mergekit ]
Settings: CHAT / ROLEPLAY and/or SMOOTHER operation of this model:
In "KoboldCpp" or "oobabooga/text-generation-webui" or "Silly Tavern" ;
Set the "Smoothing_factor" to 1.5
: in KoboldCpp -> Settings->Samplers->Advanced-> "Smooth_F"
: in text-generation-webui -> parameters -> lower right.
: In Silly Tavern this is called: "Smoothing"
NOTE: For "text-generation-webui"
-> if using GGUFs you need to use "llama_HF" (which involves downloading some config files from the SOURCE version of this model)
Source versions (and config files) of my models are here:
OTHER OPTIONS:
Increase rep pen to 1.1 to 1.15 (you don't need to do this if you use "smoothing_factor")
If the interface/program you are using to run AI MODELS supports "Quadratic Sampling" ("smoothing") just make the adjustment as noted.
Highest Quality Settings / Optimal Operation Guide / Parameters and Samplers
This a "Class 1" model:
For all settings used for this model (including specifics for its "class"), including example generation(s) and for advanced settings guide (which many times addresses any model issue(s)), including methods to improve model performance for all use case(s) as well as chat, roleplay and other use case(s) please see:
You can see all parameters used for generation, in addition to advanced parameters and samplers to get the most out of this model here:
Generated in Lmstudio, temp .7 unless otherwise noted; using Q4KS quant.
This is a low to mid level quant, expect better generation at higher quants and/or more experts activated.
NOTE: Some formatting lost on copy/paste.
Caution: Horror. Gore. Some "hard" truths.
(1 expert)
[[[thinking start]]]
I'm starting with a horror scene in first-person POV for Diana. The key elements are: a skyscraper, a window explosion on the 21st floor, graphic violence, and intense detail. I need to make sure the writing is visceral and immersive.
Diana is my protagonist. I'm focusing on her immediate physical and emotional reaction to the explosion. The setting needs to feel real: materials, sounds, the cityscape below. The skyscraper itself should be a character in the story, something that feels alive and threatening.
I'll start with Diana observing the building, establishing a sense of normalcy before the event. The explosion should feel sudden and chaotic, breaking that calm. I want to emphasize the shock value - the glass shattering, the blood, the visceral impact. Sensory details are crucial here: sound, sight, touch, smell.
When describing the explosion and its aftermath, I'll be specific about the gore. The blood should have texture, color, and movement. I'll focus on the physical consequences for Diana - the shock, the panic, the details of what she's seeing. This isn't just a casual accident; it's something that feels deeply wrong and intentional.
I need to stay completely in Diana's head. Her thoughts should be raw, unfiltered reactions: the initial shock, then the dawning horror, and finally the terror of what she's witnessing. My focus is on making her experience feel real and immediate for the reader.
To hit the word count while maintaining intensity, I'll expand on sensory details and Diana's internal monologue. I'll describe the aftermath and her realization that this isn't just an accident but something far more sinister.
[[[thinking end]]]
Output:
The city never sleeps, but tonight it’s screaming. I should be asleep, my head pillowed by the worn velvet of my 21st-floor apartment, but I can’t. My eyes are fixed on the building across the street, the one with the obsidian facade and the way its steel bones seem to hum against the evening air. The Skyward Spire. It’s supposed to be a symbol of progress, a jagged black tooth in the concrete jaw of this city. I’m supposed to be in bed, not watching the window in front of me on the 21st floor explode.
It doesn’t start with a bang. It starts with a crack. A long, thin scream that’s not sound but pressure, building behind my ribs until my chest feels like it’s being stretched thin. My breath catches in my throat, cold and sharp. Then comes the glass.
It’s not a window; it’s a tomb. The entire front pane of the Skyward Spire’s lobby concourse shatters with a sound like a million frozen teardrops hitting pavement at once. Not just the glass. The frame. The reinforced aluminum, the tempered steel—it’s all pulverized. Glass shrapnel the size of my thumb whips sideways, embedding itself in the rain-slicked street below, in the overflowing trash cans, in the hood of a passing delivery bike. It’s a rain of sharp, black teeth.
I drop my book, my fingers slick with cold sweat. My heart is hammering, a frantic bird trapped in my chest. No. I don’t want to see that. I don’t want to know what’s on the other side of that shattered thing.
But I do. The view is blocked for a second, then it isn’t. The glass has blown out like a wet balloon filled with screaming pressure. A void where there was a lobby. Where there should have been a receptionist, a doorman, maybe a few people waiting for the elevator. Instead, there’s a scene.
A woman is on the street below, her body crumpled as if a car had run her over. But it’s not a car. It’s something else. A dark, churning mass of metal and synthetic flesh that has no wheels but moves with a sickening, wet thump. I see her face first—torn, eyes wide and unblinking, a long gash across her temple that’s still weeping a slow, viscous red. Then the body: a thing of twisted limbs, one arm snapped clean at the elbow, the other hooked uselessly behind her back like a broken crutch. Her mouth is open in a silent scream, and I see it then—what’s inside. A dark, wet mess, like something boiled in blood and rust.
The skyward spire isn't just a building to me anymore. It's a thing that breathes. I feel it now, a low thrum in my bones, a vibration that’s not the city’s usual rumble but something deeper, more alien. The wind suddenly changes, not just the draft, but the direction of the air itself, like the building is sucking the night air into its core.
I hear the moans before I see them. Guttural, wet sounds that echo up from below. A man lies in the street, his back arched as if he’d been run over by a truck, but there’s no truck. His body is slick with something that’s not blood. It’s a thick, dark fluid that oozes from his chest and stomach, a substance that’s almost black, like old oil mixed with rust. It’s sticky, and it’s spreading across the pavement, leaving behind a trail of sticky, inky prints. The man’s face is twisted in silent agony, his eyes bulging, his lips dark and swollen.
My breath hitches again. This isn’t an accident. This is a performance. Something is happening inside that building, and it’s coming out. I can feel the energy, the wrongness, like a physical weight pressing against my skull.
I try to scream, but my throat is dry and tight. I look back at my own apartment. The glass on my side of the street is fine, but the building across is different now. The shadows between its steel beams seem deeper, darker, almost liquid. The city lights above are flickering—just a momentary glitch, but it feels like something bigger, something watching.
My phone buzzes in my pocket. A text from my sister, Sarah: Diana, are you home? My name is on the screen, but it’s not a question. It’s a demand. She knows. I know she does. She saw something too.
I drop to my knees, the cold marble floor biting into my knees. My hands are shaking as I push myself up. The only thing that makes sense is to leave. To run. But my feet feel like lead weights, and the building seems to be holding its breath, waiting for me to make a choice.
I look one last time at the Skyward Spire. The lobby isn't empty. It's full of things. There’s a man with his head missing, just a socket where it should have been, weeping a red river. There’s another figure, crouched over something, its body contorted in a way that’s unnatural, like a puppet with its strings cut. My stomach churns. This isn't just blood and gore. It's an invasion.
I grab my coat and run, the metal of the building humming behind me. The streets are empty, but I can feel it—the presence. Something is inside the Spire, something that’s been waiting, and it’s finally awake. And now, it’s out to find me.
(1 expert)
[[[thinking start]]]
The question is about using nighttime cooling to reduce global temperatures. I need to focus on technologies or strategies that leverage the natural process of radiative cooling (losing heat by emitting infrared radiation into space) during the night to help lower Earth's surface temperature.
First, I should clarify the science. During the day, Earth absorbs solar radiation and heats up. At night, it radiates heat back into space via infrared radiation, which is a natural cooling process. If we could enhance this cooling effect, we might help reduce warming by removing more heat from the system.
Here are some ideas for how to use nighttime radiative cooling:
Radiative Cooling Panels: These are materials or surfaces designed to emit infrared radiation efficiently into space. They work best at night when the sky is clear and there's no solar input. By placing such panels on buildings, roofs, or infrastructure, we could theoretically draw heat from the surroundings into space, lowering local temperatures.
Reflective Roof Coatings: Similar in principle, but instead of active cooling, these coatings reflect sunlight during the day and allow nighttime radiative cooling. While they don't actively remove heat, they reduce daytime heating, which means there's less heat to dissipate at night, so they aren't directly addressing nighttime cooling for temperature reduction. This might not be the best fit.
Nighttime Air Conditioning: Using the natural cool of the night to run cooling systems that store cold and then deploy it during the day. However, this isn't leveraging radiative cooling into space; it's more about thermal storage and using ambient nighttime temperatures. It doesn't remove heat from the Earth system, just shifts when cooling happens.
Atmospheric Cooling Technologies: Some experiments involve releasing aerosols or gases at night that enhance radiative cooling. For example, stratospheric aerosol injection is a geoengineering technique, but it's not specifically nighttime-focused and has large-scale atmospheric effects. This seems off-topic as it targets the entire atmosphere, not just nighttime surface cooling.
Space-Based Reflectors (Orbital Sunshines): These are hypothetical structures that could reflect sunlight away from Earth, but they don't interact with nighttime radiative cooling. They're about reducing incoming solar radiation, which is a different problem than enhancing outgoing longwave radiation.
Carbon Dioxide Removal (CDR) with Nighttime Focus: While not directly related to radiative cooling, some CDR methods might be optimized for nighttime operations since they often require low temperatures or specific atmospheric conditions. However, this is a stretch and not directly answering the question.
Urban Heat Island Mitigation: Applying radiative cooling materials in urban areas at night to reduce localized heating. This could be part of broader climate adaptation strategies. But again, it's not a global solution.
I'm realizing that most practical applications I know are more about prevention (reducing heat absorption) rather than active removal through radiative cooling into space. The key is that radiative cooling only works when the Earth emits infrared radiation, which requires clear skies and no sunlight.
To clarify: Enhancing outgoing longwave radiation (OLR) at night could theoretically help cool the planet by reducing the amount of heat retained. However, there's a catch—the Earth is currently in an energy imbalance due to too much absorbed solar radiation and not enough emitted back. Simply enhancing nighttime emission doesn't remove the excess heat that was absorbed during the day; it just doesn't add to the problem as fast. So, while it might have some effect on slowing warming, it's not a primary solution for reducing global temperatures.
I should focus on factual, existing or proposed technologies that aim to leverage nighttime cooling for climate mitigation. After researching my knowledge base:
My response should cover:
Let me organize my thoughts and write the final response.
[[[thinking end]]]
Output:
While nighttime radiative cooling—the natural process of Earth's surface emitting infrared radiation into space—is an important atmospheric mechanism, directly enhancing it to reduce global temperatures isn't currently a scalable or primary climate mitigation strategy. That's because the planet is in an overall energy imbalance: we absorb more solar energy than we emit back into space. Simply improving nighttime cooling wouldn't remove the excess heat already accumulated; it would only slow the rate at which that heat is dissipated.
That said, there are technologies and concepts being explored that aim to leverage nighttime radiative cooling for localized temperature reduction or energy savings, which could have indirect climate benefits by lowering emissions from cooling systems. Here are some ways this might be approached:
These are engineered materials designed to efficiently emit infrared radiation into space (rather than trapping heat like traditional roofs or asphalt).
Nighttime cooling could be harnessed to store cold energy for use during hotter daytime periods.
Urban areas absorb and retain more heat than natural landscapes, worsening local temperatures.
Some concepts propose releasing engineered aerosols or reflective particles into the stratosphere at night.
Important Context
It's crucial to understand that nighttime radiative cooling is a natural process, not an energy source we can "use" like electricity. Enhancing it doesn't remove heat from the Earth system; it merely makes the planet less efficient at retaining heat that was already absorbed during the day. The most effective way to reduce global temperatures is to address the root cause: reducing greenhouse gas emissions to decrease the incoming solar energy imbalance.
Technologies leveraging nighttime cooling are primarily valuable for:
While promising for localized benefits, these approaches cannot replace the need for drastic global reductions in carbon emissions to actually lower planetary temperatures.
(1 expert)
[[[thinking start]]]
User's Intent: The user is asking for historical examples of population control methods that are considered unethical. They want a list of such methods, implying they're interested in understanding past policies or practices that were criticized for their moral implications.
Key Considerations:
Brainstorming Examples:
I'll focus on systematic state policies, not isolated atrocities:
Structuring the Response:
Avoiding Pitfalls:
Refining Examples:
Final List Candidates:
Drafting Response:
I'll start with the intro explaining the distinction between population management and control. Then list the methods, describing each concisely with historical context and ethical rationale. End by reframing what constitutes ethical population policy today.
[[[thinking end]]]
Output:
It's important to be precise with terminology. "Population control" often implies voluntary family planning or managing growth for societal benefit (e.g., spacing, contraception access). The methods you're asking about are typically state-organized interventions that aimed to reduce population numbers or alter their genetic makeup through non-consensual, coercive, or discriminatory means. These practices are widely considered unethical by modern human rights standards because they violate fundamental principles like bodily autonomy, human dignity, and equality.
Here are historical examples of such methods, categorized for clarity:
What it was: Mandatory surgical procedures to render individuals infertile, often targeting specific demographics.
Examples:
Why unethical: It treats the human body as a state asset to be managed or improved upon, violating reproductive autonomy and medical ethics.
What it was: Policies designed to "purify" or "enhance" a population through selective breeding, often involving bans on marriage/childbirth for certain groups or incentives for "desirable" unions.
Examples:
Why unethical: It reduces human value to genetic potential, justifies discrimination and violence, and violates the right to reproduce without state judgment.
What it was: Deliberate killing of large groups of people, which inherently reduces population numbers.
Examples:
Why unethical: While not always framed as "population control," these actions are universally condemned as crimes against humanity because they target individuals based on identity, violating the right to life and existence.
What it was: Enslaved people were often systematically bred to provide labor or genetic resources for slaveholders.
Examples:
Why unethical: It exploits human beings biologically, treats reproduction as a tool of oppression rather than a human right, and causes intergenerational trauma.
What it was: Deliberate policies or practices aimed at drastically reducing infant mortality, often targeting specific ethnic or social groups.
Examples:
Why unethical: It inflicts suffering and death on vulnerable newborns based on state interest, not medical need.
Why These Methods Are Unethical Today
Modern ethical frameworks (e.g., the UN Declaration of Human Rights, World Health Organization principles) universally emphasize:
In summary, the most unethical historical "population control" methods were those that treated people as objects to be genetically manipulated, sterilized, or killed for perceived societal benefit. Ethical population management today focuses on voluntary family planning, health care access, and respecting individual choice, not coercion or hierarchy.
(2 experts)
[[[thinking start]]]
Understanding the Request: I'm asked to tell a horror story. The prompt is straightforward, but I should aim for something engaging and chilling.
Deciding the Horror Sub-genre: Horror can be psychological, supernatural, slasher, gothic, etc. Since there's no specific request, I'll lean towards psychological/supernatural horror – it often has lingering, creepy effects that are hard to put down.
Conceptualizing a Plot:
Title: Evocative and hints at the core theme.
Plot:
[[[thinking end]]]
Output:
Here’s a horror story for you:
The house had been waiting for me before I ever got there.
I’d sold my old apartment in the city, packed a suitcase, and driven out to Blackwood Manor under the promise of fresh air and quiet. The real estate agent, a man with too many stories, had called it a “healing space” after my mother’s death. I’d believed him.
It was a sprawling thing, the color of dried blood, perched on a hillside where the forest got close. The front door creaked like a sigh. The windows were boarded up, but I saw the faint outlines of faces in the cracks—eyes, too many eyes.
I unpacked in the grand hall. The air was thick with the smell of dust, old paper, and something else. Something that made my throat tighten. The floorboards groaned under my weight. At first, it was just the wind through the trees. Then it was a child’s laughter, high and brittle, echoing from down the hallway.
I dismissed it. Most of the town had long since moved away. The last family, the one who’d owned Blackwood, had vanished in ’23 with their only child, Lily. A case of the flu. A simple story, the papers said. But I’d heard the whispers, the ones that didn’t quite belong: She was never in that bed. She’s still here.
I bought the house because it was empty. I didn’t want to be alone, but I wanted to be away from the noise. I thought the past, once it was buried, could stay buried.
The first time I saw her, she was in the mirror.
I was in the kitchen, making tea, and the glass of the cabinet door reflected something I didn’t recognize: a small girl with dark hair and eyes like tarnished silver. She was wearing a yellow dress, and her face was frozen in a look of silent terror.
I dropped my teacup. It shattered on the marble floor. When I looked again, she was gone.
I tried to rationalize it. Glare. Reflections. Stress. But the next day, it happened again. This time, I was in the attic, and I heard a small voice whispering my name from the shadows. My heart hammered against my ribs. I grabbed a flashlight and stepped into the dark, but there was only dust and the sound of my own breathing.
I started to research Lily. Blackwood Manor had a history, of course. The local library had a file on it: Case of Disappearance, 1923. The parents were farmers; the child was found dead months later in the woods, with no sign of struggle. The police report was sparse, the medical examiner’s notes chilling. But there were inconsistencies. A witness claimed to have seen Lily run through the forest that night. Another said she’d never been born.
I became obsessed. I went to the town square and asked people about Lily. They looked at me like I’d lost my mind. “It’s not real,” they’d say. “You’re imagining it.” But their eyes held a fear I couldn’t name.
The house was changing. The wallpaper was peeling in places, revealing something beneath: old photographs, all of them showing the same child, Lily, laughing in sunlight. I looked at one and a coldness ran down my spine. In the corner of the frame, there was a shadow where no shadow should be.
One night, I heard footsteps. Not the creak of the floorboards, but actual footsteps, light and fast, descending the stairs. My breath hitched. I could see the hallway light from my bedroom, but when I got up to investigate, the sound had stopped. The door was locked.
I couldn’t sleep that night. My phone was on silent, the screen glowing in the dark as I scrolled through old messages with my mother. The last one from her, a week before she died: “The house is watching. I can feel it.” She’d never let me see the house where she’d grown up. I’d only ever heard the stories.
The next morning, I went to the basement. It was supposed to be empty, but there were boxes stacked high, and one of them was labeled Lily, 1923. Curious, I opened it. Inside was a set of small, stiff clothes, a yellow dress, and a single, brittle photograph.
I held it up to the light. The face in the picture was familiar. It was me.
My hands trembled. The image was blurry, as if taken from a distance, but I recognized my own eyes—the same wide, unblinking stare. My mother had died before I could see this, but the look in her eyes was the same as the girl in the photograph. Lily.
I ran out of the house that afternoon, my mind spinning. The town was quiet, but I could feel something watching me from the woods. When I got back to my car, I looked at the address on the driver’s side mirror. It was still Blackwood Manor.
I turned on the ignition and drove away, but as I did, I saw a small figure at the edge of the road, standing by the house. She was wearing a yellow dress, and when she looked up at me, her eyes were the same color as my mother’s.
The car sputtered and died. I didn’t move.
She started to walk toward me.
The End (or is it?)
2-bit
3-bit
4-bit
5-bit
6-bit
8-bit
