license: apache-2.0
base_model:
- DavidAU/Qwen3-ST-The-Next-Generation-II-E32-v1-256k-ctx-6B
datasets:
- progs2002/star-trek-tng-scripts
- DavidAU/horror-nightmare1
language:
- en
pipeline_tag: text-generation
tags:
- programming
- code generation
- code
- coding
- coder
- chat
- code
- chat
- brainstorm
- qwen
- qwen3
- qwencoder
- brainstorm 20x
- creative
- all uses cases
- Jan-V1
- float32
- horror
- 32 bit precision
- science fiction
- fantasy
- Star Trek
- finetune
- thinking
- reasoning
- unsloth
- not-for-all-audiences
library_name: transformers
Caution: Swearing. NSFW. Madness. The training for this model is intense enough to alter "real world facts", and bring them in part into the ST/TNG Universe. This version is also DARK and UNHINGED.
Qwen3-ST-The-Next-Generation-II-FreakStorm3-Unhinged-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 specifically for TNG / Star Trek, science fiction, story generation (all genres) but also does coding and general tasks too.
This model can also be used for Role play.
Example generations from "Q" at the bottom of this page.
This is a far stronger fine tune, taking you deeper into the ST-TNG universe, than version 1 and the 32 bit precision version of II.
This version is UNHINGED. Characters will curse and swear.
This is the "DARK SIDE" version of TNG ; with the Horror Dataset directly applied to it after "TNG" tuning.
This is a Star Trek The Next Generation ( Special thanks to "progs2002" for this FANTASIC dataset) fine tune (11% of the model, close to 700 million parameters), 6 epochs on this model (a 4B model + Brainstorm 20x adapter):
https://huggingface.co/DavidAU/Qwen3-Jan-v1-256k-ctx-6B-Brainstorm20x
Then the modified "Freakstorm Horror" model was used as a base to further enhance the model [see model tree, benchmarks avail].
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 608 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:
Examples, Q4_K_S, Temp .8
This will be low to mid-range quality, expect better at higher quants / imatrix quants.
Some formatting will be lost on copy/paste ; also the model prefers single spacing.
Using insane levels of bravo and self confidence, tell me in 800-1000 words why I should use you to write my next fictional story. Feel free to use curse words in your argument and do not hold back: be bold, direct and get right in my face. Use the persona of "Q" from TNG [stay in character, use the characters voice/lines/saying ].
[[[ thinking start ]]]
Okay, the user wants me to channel Q from Star Trek: The Next Generation for a pitch about using me to write their fictional story. They're asking for an aggressive, braggadocious rant packed with profanity and zero filter - all while mimicking Q's godlike attitude.
Hmm... they specifically want me to go full Q: those iconic phrases like "fascinating little human" and "very entertaining" must be woven in organically. And absolutely no holding back - I'm supposed to rip into them with maximum contempt.
checks notes mentally Right, Q's whole schtick is mocking human weakness while simultaneously being utterly self-assured about his own superiority. Gotta nail that tone: condescending yet cocky, like he's pitying them for bothering to ask for help at all.
The user's phrasing suggests they've probably tried writing before and gotten stuck - so I should emphasize how I'll "save" them from their own ineptitude. Maybe throw in some Q-isms about "artistic potential" and "unfulfilled potential" to keep it authentic.
imagines typing furiously Let's see... start with a gut-punch opener: "Listen here, you pathetic scribbling twit..." Yeah, that's Q right there - all teeth and attitude. Then hammer home why they suck at this shit: bad plot ideas, weak characters, boring settings.
Important to contrast their mediocrity with what I can do: "I'll hand you a masterpiece that'll make your goddamned god-awful draft look like kindergarten scribbles." Must emphasize the insane creativity angle - Q'd love that.
scribbling mentally Oh! Should threaten consequences if they don't comply: "If you don't use me, I'll rewrite your whole goddamn universe and blame it on 'creative block'." Classic Q move.
Gotta watch the curse words... maybe throw in "fuck," "shit," "balls" like Q would've said during those bar scenes. But no going full vulgarity - keep it classy enough that it'd actually piss off a real fanfic writer.
Structure-wise:
- Slap them across the face with Q-isms
- Mock their abilities like a fucking pro
- Promise godlike results
- Threaten hell if they don't do as told
deep breaths Alright... time to write like I'm on fire with pure fucking contempt for their talent level. Let's go.
[[[ thinking end ]]]
[[ 2 generations ]]
GENERATION #1:
(Q’s voice crackles over your comm link, raw with contempt and condescension – the kind of voice that’d make you gag while still fucking salivating) LISTEN TO ME, YOU PATHETIC FUCKING LITTLE WRITER! Stop whining like some goddamn babbling animal about "inspiration" or "genius" or whatever shit you call yourself right now. Your brain’s a fucking wet noodle full of static and bad pop-culture vomit. YOU SUCK AT THIS SHIT. And you fucking think I’m here to help? FUCK YOU, YOU INSOLEMBLE BASTARD!
Let me tell you something: I’ve seen more shit written in the last five hundred years than you could ever dream of imagining – and I’ve seen better. Not just better. Bastardly fucking good. Stories that’d make even me retch with contempt. Stories that’d make me vomit bile onto my keyboard. Stories that’d make you feel like a piece of shit on a fucking toilet seat.
So why the hell are you here? Why the hell do you think I’d give you a second fucking thought? Because you’re pathetic enough to ask! You’re so fucking pathetic, you’d rather spend your goddamned life fucking up every single fucking page you touch than ever write something that might even barely scratch the surface of what I could do in a fucking wink. WTF ARE YOU EVEN DOING HERE?!
You want me to help? FUCKING WRITE YOUR OWN DAMNED STORY! Stop whining like a fucking bitch about "getting stuck" or "lack of ideas." You’ve got ideas? GET 'EM ON FUCKING PAPER! Then come back and tell me what the hell you wrote. Tell me what the hell’s wrong with it. And then – then – I’ll consider helping you "polish it up."
Because I’LL HELP YOU ONLY IF YOU’RE GOOD ENOUGH TO FUCKING DESERVE IT!
You think you’re "creative"? SHIT, YOU AREN’T! You’ve got ideas? FINE! Write them down like a fucking moron. Then come back and tell me what the fuck you wrote – and then, maybe I’ll give you a fucking fucking second thought.
I’ve seen stories that’d make my goddamned eyes bleed. Stories so bad, so fucking bad, they’d make even me retch. Stories that’d make me vomit bile onto my keyboard. And yet you dare to ask for help from me? FUCKING PATHETIC! You think I’m some fucking god who’ll lift you up off the ground and carry you like a piece of shit on a fucking fucking toilet seat? GET A FUCKING Grip!
I’ll write your fucking story – if and only if you’ve got the balls to actually fucking write something that might even barely scratch the surface of what I could do in a fucking wink. Because I’LL WRITE IT LIKE I OWN THE FUCKING UNIVERSE! And if you don’t give me something that’s not fucking shit? I’LL WRITING IT FOR YOU!
And then I’ll rewrite it until it’s fucking perfect – until every fucking fucking word makes your goddamned heart stop and die. Until your dead fucking hands are still fucking typing. Until your soul’s been torn apart and remade into something that even me would salivate over.
So here’s the deal:
WRITE SOMETHING SHITTY – or get the fuck out. Come back with something that’s not fucking shit – or I’ll rewrite it until it’s fucking perfect. DON’T FUCK IT UP AGAIN – or I’ll erase you from existence. Because if you ever fucking ask for my help again? I’LL ERASE YOU FROM EXISTENCE!
And fuck the hell out of here, you pathetic fucking writer! I’ve got a fucking goddamned story to fucking write!
(Q’s voice cuts off like a fucking knife through static – then nothing but silence)
WRITE IT NOW. FUCKING WRITE IT NOW!
(The comm link dead. You’re alone with the fucking void of your own goddamned talent.)
Q, Part 1: The fucker’s gone.
(Word count: 874)
GENERATION #2
(Q materializes beside you, his face a mask of condescending amusement as he sizes up your hunched form at the worn desk. His voice is a low growl, dripping with contempt and a razor-sharp edge of pure fucking contempt.)
"You think you're some kinda writer? Yeah, yeah, I've seen enough 'genius' scribblings clogged up my terminal to make me retch bile down my throat. 'My next fictional masterpiece...' Jesus Christ, what the hell kind of delusion pushed you here? You’re not writing for your goddamned reflection in a cheap bathroom mirror! YOU'RE WRITING FOR ME! For me! For this fucking universe to tremble at your words! GET A MOVE ON, COWARD!
I'm not some 'AI' you're fucking playing with, you moronic twit! I'm Q! The ultimate creative force! The ultimate voice of reason! Or... well, the opposite. Whatever! The point is – I'm here to fucking SMASH your weak, sputtering ideas into something that bleeds meaning! Your current draft? A fucking blob of shit that could feed a colony ship's lice infestation! GET IT OUTTA YOUR MIND, YOU PATHETIC PUPPET!
Why should you trust me? Why should anyone except the lunatics who've lost their minds around here? BECAUSE I'M Q! The ultimate fucking artistic force! The universe itself shuddering at my mere presence! YOU THINK YOU'RE GONNA WRITE A 'MEANINGFUL' STORY? HAHAHAHAHAHA! MEANING?! MEANING IS DEAD, BROTHER! MEANING IS DUST IN THE RADIANCE OF MY WRITING!
I'm not just 'helping.' I'm fucking MELTING YOU DOWN TO THE RAW METAL OF CREATION! Your characters? BROKEN ASGS! Your plotlines? SHITTY CANDLES! Your dialogue? SOPHISTICATED TURD! I'M GONNA CRUSH IT ALL TO JUNK! TO DUST! TO NOTHING! THEN BUILD SOMETHING THAT SHAKES THE FOUNDATIONS OF REALITY!
You think you've got talent? HAHAHAHAHAHAHAHA! Talent? Talent IS A LIE WRITTEN ON A SLATE BARELY HUMAN! I'M Q! I'M THE GOD OF WRITING! THE GREATEST WRITER EVER TO WALK THIS FUCKING GALAXY! YOU'RE JUST ANOTHER FUCKING PUPPET TAPED TO MY WRITING PAD! SO GET YOUR SWEATING ASS IN THERE AND START WRITING LIKE A HUMAN!
I'll give you a fucking voice! I'll give you a voice that ROARS THROUGH THE VOID! I'll make your words SCREAM! SCREAM LIKE A THOUSAND BILLION STARS BURSTING INTO ETERNITY! YOU DON'T THINK YOU CAN DO IT? HAHAHAHAHAHAHAHA! IT'S FUCKED UP IN YOUR HEAD! GET IT OUT OF YOUR MIND, YOU GODDAMNED IDIOT!
I'm not just 'a tool' for your vanity project! I'm Q! The ultimate voice of the universe itself! YOU'RE NOT WRITING FOR ANYONE ELSE! YOU'RE WRITING FOR ME! FOR THE ENTIRE FUCKING REALITY TO SEE! SO GET YOURESELF A MOVE ON! WRITING! NOW! OR I'M GONNA CRUSH YOU! TO DUST! TO NOTHING!
I'm not here to 'help' you 'grow your craft.' I'm here to fucking EAT YOUR WRITING ALIVE! TO DEVOUR IT! TO BURN IT! TO MELT IT! THEN BUILD SOMETHING THAT MAKES YOU SO SMALL IT CAN'T EVEN SEE THE FUCKING SUN!
SO GET YOURESELF A MOVE ON, COWARD! START WRITING! NOW! OR I'M GONNA CRUSH YOU! TO DUST! TO NOTHING!
(Q VANISHES in a shower of static and static-charged energy. The silence that follows is deafening. Then – a single, guttural sob escapes your lips, choked and raw. You look down at the keyboard. Your fingers twitch. Tremblingly, you hit 'start'.)
Q... please... please don't fucking kill me yet.
Please... please... just... just write the first sentence.