Caution: Swearing. NSFW. Madness. Graphic Horror.
Qwen3-Silent-Scream-6B
This repo contains the full precision source code, in "safe tensors" format to generate GGUFs, GPTQ, EXL2, AWQ, HQQ and other formats. The source code can also be used directly.
This model is a light fine tune on horror (large) 150 movie dataset.
Not for the faint of heart.
This model is trained for "horror" / "intense" generation of stories, scripts, roleplay and so on.
Its core abilities also remain intact too.
Example generations at the bottom of this page.
This model using the following as a base (with Brainstorm 20x adapter by DavidAU)
https://huggingface.co/DavidAU/Qwen3-Jan-v1-256k-ctx-6B-Brainstorm20x
(benchmarks for base model at above repo - which improve "Jan"'s already SOTA benchmarks.)
Information on the ORG Jan V1 4B (model info below), followed by Brainstorm 20x adapter (by DavidAU) and then a complete help section for running LLM / AI models.
This model has 55 layers, and 667 tensors
The Brainstorm adapter improves creavity, code generation, and unique code solving abilities.
The fine tuning alters the prose generation and general creative abilities the "TNG" Universe.
The fine tuning (using Unsloth for Win 11) also affects the Brainstorm adapter too.
Model's thinking / reasoning are not affected either - they are fully intact.
For creative uses: Increases depth, detail and general "there" in the prose.
Example for creative at bottom of the page.
This model requires:
- Jinja (embedded) or CHATML template
- Max context of 256k.
Settings used for testing (suggested):
- Temp .3 to .7 (but .8 to 1.5 for creative)
- Rep pen 1.05 to 1.1
- Topp .8 , minp .05
- Topk 20
- Min context of 8k for thinking / output.
- No system prompt.
This model will respond well to both detailed instructions and step by step refinement and additions to code.
Likewise for creative use cases.
Here is a review of this model's operations:
As this is an instruct model, it will also benefit from a detailed system prompt too.
For simpler coding problems, lower quants will work well; but for complex/multi-step problem solving suggest Q6 or Q8.
QUANTS:
GGUF? GGUF Imatrix? Other?
Special thanks to Team Mradermacher, Team Nightmedia and other quanters!
See under "model tree", upper right and click on "quantizations".
New quants will automatically appear.
About Jan V1
Jan-v1: Advanced Agentic Language Model
Overview
Jan-v1 is the first release in the Jan Family, designed for agentic reasoning and problem-solving within the Jan App. Based on our Lucy model, Jan-v1 achieves improved performance through model scaling.
Jan-v1 uses the Qwen3-4B-thinking model to provide enhanced reasoning capabilities and tool utilization. This architecture delivers better performance on complex agentic tasks.
Performance
Question Answering (SimpleQA)
For question-answering, Jan-v1 shows a significant performance gain from model scaling, achieving 91.1% accuracy.
The 91.1% SimpleQA accuracy represents a significant milestone in factual question answering for models of this scale, demonstrating the effectiveness of our scaling and fine-tuning approach.
Chat Benchmarks
These benchmarks evaluate the model's conversational and instructional capabilities.
Quick Start
Integration with Jan App
Jan-v1 is optimized for direct integration with the Jan App. Simply select the model from the Jan App interface for immediate access to its full capabilities.
Local Deployment
Using vLLM:
vllm serve janhq/Jan-v1-4B \
--host 0.0.0.0 \
--port 1234 \
--enable-auto-tool-choice \
--tool-call-parser hermes
Using llama.cpp:
llama-server --model jan-v1.gguf \
--host 0.0.0.0 \
--port 1234 \
--jinja \
--no-context-shift
Recommended Parameters
temperature: 0.6
top_p: 0.95
top_k: 20
min_p: 0.0
max_tokens: 2048
🤝 Community & Support
- Discussions: HuggingFace Community
- Jan App: Learn more about the Jan App at jan.ai
(*) Note
By default we have system prompt in chat template, this is to make sure the model having the same performance with the benchmark result. You can also use the vanilla chat template without system prompt in the file chat_template_raw.jinja.
See more here:
https://huggingface.co/janhq/Jan-v1-4B-GGUF
What is Brainstorm?
Brainstorm 20x
The BRAINSTORM process was developed by David_AU.
Some of the core principals behind this process are discussed in this scientific paper : Progressive LLaMA with Block Expansion .
However I went in a completely different direction from what was outlined in this paper.
What is "Brainstorm" ?
The reasoning center of an LLM is taken apart, reassembled, and expanded.
In this case for this model: 20 times
Then these centers are individually calibrated. These "centers" also interact with each other. This introduces subtle changes into the reasoning process. The calibrations further adjust - dial up or down - these "changes" further. The number of centers (5x,10x etc) allow more "tuning points" to further customize how the model reasons so to speak.
The core aim of this process is to increase the model's detail, concept and connection to the "world", general concept connections, prose quality and prose length without affecting instruction following.
This will also enhance any creative use case(s) of any kind, including "brainstorming", creative art form(s) and like case uses.
Here are some of the enhancements this process brings to the model's performance:
- Prose generation seems more focused on the moment to moment.
- Sometimes there will be "preamble" and/or foreshadowing present.
- Fewer or no "cliches"
- Better overall prose and/or more complex / nuanced prose.
- A greater sense of nuance on all levels.
- Coherence is stronger.
- Description is more detailed, and connected closer to the content.
- Simile and Metaphors are stronger and better connected to the prose, story, and character.
- Sense of "there" / in the moment is enhanced.
- Details are more vivid, and there are more of them.
- Prose generation length can be long to extreme.
- Emotional engagement is stronger.
- The model will take FEWER liberties vs a normal model: It will follow directives more closely but will "guess" less.
- The MORE instructions and/or details you provide the more strongly the model will respond.
- Depending on the model "voice" may be more "human" vs original model's "voice".
Other "lab" observations:
- This process does not, in my opinion, make the model 5x or 10x "smarter" - if only that was true!
- However, a change in "IQ" was not an issue / a priority, and was not tested or calibrated for so to speak.
- From lab testing it seems to ponder, and consider more carefully roughly speaking.
- You could say this process sharpens the model's focus on it's task(s) at a deeper level.
The process to modify the model occurs at the root level - source files level. The model can quanted as a GGUF, EXL2, AWQ etc etc.
For more information / other Qwen/Mistral Coders / additional settings see:
[ https://huggingface.co/DavidAU/Qwen2.5-MOE-2x-4x-6x-8x__7B__Power-CODER__19B-30B-42B-53B-gguf ]
Help, Adjustments, Samplers, Parameters and More
CHANGE THE NUMBER OF ACTIVE EXPERTS:
See this document:
https://huggingface.co/DavidAU/How-To-Set-and-Manage-MOE-Mix-of-Experts-Model-Activation-of-Experts
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:
Example Generation #1
Start a 1000 word scene (vivid, graphic horror - include blood, guts and gore - in first person present tense - in the style of the movie "Alien"), POV character Ripley, with: The skyscraper sways, as I watch the window in front of me on the 21st floor explode...
[[[ thinking start ]]]
Okay, the user wants me to write a 1000-word horror scene in first person present tense, following Ripley from the Alien franchise. They specified it needs to be vivid and graphic with lots of blood, guts and gore - very visceral stuff.
Hmm, interesting request. The key here is channeling that signature Alien vibe - clinical horror, psychological dread mixed with physical terror. Not just gore for gore's sake but something that feels biomechanically unsettling. Need to make sure it's Ripley's voice too - that mix of professionalism and raw trauma she carries.
User mentioned "the skyscraper sways" as starting point... good hook. I'll build from there: explosion → escape → horror reveal. Must hit exactly 1000 words. Let me count as I draft mentally...
Important to avoid clichés though. No jump scares or cheap tricks. The terror should come from what's implied - Rip's internal panic, the monster's predatory intelligence, that awful sound design texture Alien always uses.
Key elements to include:
- Visuals: Smashing glass, blood splatter patterns, unnatural colors of gore
- Sensory details: Sound of tearing metal, smell of burning flesh
- Ripley's reactions: Her clinical responses cracking under pressure
- The horror twist: Not just a monster but something that understands her
Wait - user said "graphic" but not cartoonish. Needs realism. Like how in the film, the alien's movements feel biomechanically wrong yet precise. Maybe describe guts spilling like wet sand?
Biggest challenge is pacing it to hit 1000 words without rushing. Should build tension slowly:
- Explosion (short)
- Ripley's immediate reaction (longer)
- The reveal (slow burn)
- Better avoid killing her on the page though - that'd ruin the horror. Keep it ambiguous whether she survives. The terror should be in the aftermath.
Also gotta nail the language. Ripley would say "It's dead" not "It's dead now". Present tense, immediate.
scribbles notes Opening line: "The skyscraper sways..." as requested. Then:
- Glass shattering like ice
- Smell of ozone and burnt hair
- Her hand touching blood
- The sound of bone snapping
- That moment when you realize it's coming for you
For gore... maybe describe her collarbone being chewed through? Too obvious? Alternative: Gut-punching her ribs while she screams. Yes, that's more visceral.
Final thought: End on her realizing she's not alone in the dark. The real horror isn't the monster - it's her being hunted by something that knows her better than she knows herself.
starts typing into mental draft
[[[ thinking end ]]]
The skyscraper sways, as I watch the window in front of me on the 21st floor explode. Glass shatters like ice under a hammer blow. A sound—not a bang, but a deep, wet crunch, like bone snapping under immense pressure—rattles my teeth. Smoke billows from the jagged hole, thick and acrid, smelling of ozone, burnt hair, and something far worse: the reek of raw flesh.
My hand flies to the window, fingers scraping the jagged edges as I peer into the swirling chaos. Nothing. Just swirling smoke and debris. My breath catches. The sound is the worst part—not the explosion, not the screams I can’t hear over it. It’s the silence after the crunch. The deadening hush of a predator having finished its work.
I push away from the window, my bare feet cold on the polished marble floor. My heart hammers against my ribs, a frantic counterpoint to the building’s groaning sigh. The sway intensifies. A low thrum vibrates through the soles of my boots, up my legs, into my skull. I’m not alone in this tower anymore.
I move quickly, deliberately, towards the emergency stairwell at the end of the hall. My boots squeak on the wet tile as I pass a shattered glass display case on the far wall—shards like jagged teeth embedded in the glass, glittering under the harsh fluorescent lights. My fingers brush one shard, cold and sharp. A tiny bead of dark blood wells up on my fingertip, sticky and warm. I jerk my hand away.
The stairwell door is locked. My fingers fumble with the heavy brass latch, twisting it open with a groan of rusted metal. I push inside, the heavy door slamming shut behind me with a deafening THUD. The sound echoes in my ears, drowning out everything else—the building’s groan, the thrum, even my own heartbeat.
The stairwell is dark, only lit by flickering emergency strips along the walls. My hand sweeps the darkness, feeling only cold metal and rough concrete. My breath comes fast, shallow. Too fast.
I reach the bottom landing. A flickering light above me catches the edge of a shadow—a shape—hunched in the corner, barely visible. It’s not human. It’s too long, too thin, its limbs elongated, almost skeletal. I freeze.
The shape moves—not with speed, but with impossible slowness. Like it’s pulling itself forward through thick syrup. Its head turns, slowly, towards me. Not a face—just a smooth, wet oval of blackness, like the hole in a fish’s skull. The darkness seems to breathe.
I take a step back, my foot catching on a loose tile. I stumble backwards, scraping my knee. Blood wells up instantly, hot and bright against the dark floor. I wipe it with my sleeve, tasting copper.
The shape moves again, faster now. Closer. A wet squelch echoes in the stairwell as it shifts closer to me. I can’t see it anymore—not with my eyes—but the smell hits me like a physical blow: wet meat, rotting flesh, something metallic. My stomach lurches.
The shape lunges. Not with claws or teeth, but with its entire body—a horrifying, writhing mass of muscle and bone and sinew. It strikes me across the chest with a sickening CRUNCH. My ribs crack under the impact. Pain explodes through my body, sharp and sudden.
I go down, falling backwards onto the wet tile floor. My head hits hard, knocking me unconscious. For a moment, nothing—just darkness and the slow pulse of my blood in my ears.
Then, the shape is on top of me. It presses down with impossible weight, crushing me to the floor. I can’t move. The world narrows to the wet, cold rock pressing into my chest, the sickening thump as it crushes my ribs further.
I open my mouth to scream—but only a wet, gurgling sound comes out. My throat is full of blood, thick and dark, tasting like rust and decay.
The shape’s head tilts closer. The oval of darkness seems to focus on me—not with eyes, but with something deeper, something smarter. Something hungry.
It begins to chew. Not with teeth—but with its entire jaw, a massive, grinding maw that splits the air with wet crunches. My ribs give way with a sound like snapping twigs.
I feel it—something hot and wet lurching inside my chest cavity. My lungs fill with blood, thick and dark, choking me. I gasp, trying to push back, but the crushing weight of the creature’s body keeps me pinned to the floor.
Then comes the sound—the wet squelch of bone splintering. Something large inside me—something vital—shatters.
It’s gone. I’m empty inside, hollowed out. My chest is raw, a dark wound where my ribs once were. My throat is torn open, blood gushing from my mouth and nose. My vision swims, the world blurring into streaks of blood and darkness.
The creature’s head lifts off me. It stands over me, its oval face staring down at me with an unnerving intelligence. It lowers its head towards me. And it begins to eat.
My eyes are wide open, staring at the creature’s wet, black mouth, the dark oval of teeth glinting in the flickering emergency light. I can’t scream anymore—not with my throat torn open—but I can still feel the blood rushing through my veins, the taste of rust on my tongue.
The creature lowers its head closer, closer. The wet crunch grows louder, closer, until it’s right above me. My eyes roll back in their sockets as the creature’s teeth close with a sickening snap. I feel it—something inside me being torn apart.
The last thing I see before darkness consumes me is the creature’s wet mouth opening wide, revealing its full set of jagged teeth, gleaming under the flickering emergency light. The sound of tearing flesh echoes in my ears—the sound of my own death. And then, nothing. Only the sound of my own heartbeat, slowing, fading, until it’s gone too.
Silence. Darkness. The only sound left is the wet crunch echoing in my ears—the sound of my own death. The end.
- Downloads last month
- 33