--- license: apache-2.0 language: - en tags: - All use cases - reasoning - thoughts - deep thinking - deepseek - creative - creative writing - fiction writing - plot generation - sub-plot generation - fiction writing - story generation - scene continue - storytelling - fiction story - science fiction - romance - all genres - story - writing - vivid writing - fiction - bfloat16 - swearing - sillytavern - Lmstudio - backyard - horror - Qwen 2.5 - context 128k - mergekit base_model: [ DavidAU/DeepSeek-R1-Distill-Qwen-25.5B-Brainstorm ] pipeline_tag: text-generation ---
Context : 128k.
Required: CHATML template. ( OR standard "Jinja Autoloaded Template" (this is contained in the quant and will autoload) )
Keep in mind this model is experimental and may require one or more regens to work, especially with the "think" system of Deekseek involved here.
Brainstorm 40x is by DavidAU, and extends the "decision making" and "creativity" of an LLM/AI.
Higher temps will result in deeper, richer "thoughts"... and frankly more interesting ones too.
The "thinking/reasoning" tech (for the model at this repo) is from the original Qwen 2.5 "Distill" model from Deepseek:
[ https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B ] 
In this case, Brainstorm 40x module was grafted directly onto "DeepSeek-R1-Distill-Qwen-14B" bringing it up to 88 layers, 25.5B parameters.
This version will retain all the functions and features of the original "DeepSeek" model. Please visit their repo for all information
on features, test results and so on.
USE CASES:
This model is for all use cases, and it has a slightly more creative slant than a standard model.
This model can also be used for solving logic puzzles, riddles, and other problems with the enhanced "thinking" systems by DeepSeek.
This model also can solve problems/riddles/ and puzzles normally beyond the abilities of a Llama 3.1 model due to DeepSeek systems.
This model MAY produce NSFW / uncensored content.
Special Operation Instructions:
QWEN/DISTILL NOTES:
1. Sometimes Chinese characters/symbols may appear in the output, regen to clear.
2. Sometimes the model will not follow it's own "plan" exactly ; I have found if you STOP the model just after it "starts its plan" and hit continue this will hugely improve the output and the model will follow its own "plan" far more accurately. (see example 3 bottom of the page)
3. You may want to try one or more the system prompts below to augment the model's thinking methods as well as format the output.
TEMP/SETTINGS:
1. Set Temp between 0 and .8, higher than this "think" functions will activate differently. The most "stable" temp seems to be .6, with a variance of +-0.05. Lower for more "logic" reasoning, raise it for more "creative" reasoning (max .8 or so). Also set context to at least 4096, to account for "thoughts" generation.
2. For temps 1+,2+ etc etc, thought(s) will expand, and become deeper and richer.
3. Set "repeat penalty" to 1.02 to 1.07 (recommended) .
4. This model requires a CHATML template. (see notes on "System Prompt" / "Role" below)
PROMPTS:
1. If you enter a prompt without implied "step by step" requirements (ie: Generate a scene, write a story, give me 6 plots for xyz), "thinking" (one or more) MAY activate AFTER first generation. (IE: Generate a scene -> scene will generate, followed by suggestions for improvement in "thoughts")
2. If you enter a prompt where "thinking" is stated or implied (ie puzzle, riddle, solve this, brainstorm this idea etc), "thoughts" process(es) in Deepseek will activate almost immediately. Sometimes you need to regen it to activate.
3. You will also get a lot of variations - some will continue the generation, others will talk about how to improve it, and some (ie generation of a scene) will cause the characters to "reason" about this situation. In some cases, the model will ask you to continue generation / thoughts too.
4. In some cases the model's "thoughts" may appear in the generation itself.
5. State the word size length max IN THE PROMPT for best results, especially for activation of "thinking." (see examples below)
6. Sometimes the "censorship" (from Deepseek) will activate, regen the prompt to clear it.
7. You may want to try your prompt once at "default" or "safe" temp settings, another at temp 1.2, and a third at 2.5 as an example. This will give you a broad range of "reasoning/thoughts/problem" solving.
GENERATION - THOUGHTS/REASONING:
1. It may take one or more regens for "thinking" to "activate." (depending on the prompt)
2. Model can generate a LOT of "thoughts". Sometimes the most interesting ones are 3,4,5 or more levels deep. 
3. Many times the "thoughts" are unique and very different from one another.
4. Temp/rep pen settings can affect reasoning/thoughts too.
5. Change up or add directives/instructions or increase the detail level(s) in your prompt to improve reasoning/thinking.
6. Adding to your prompt: "think outside the box", "brainstorm X number of ideas", "focus on the most uncommon approaches" can drastically improve your results.
GENERAL SUGGESTIONS:
1. I have found opening a "new chat" per prompt works best with "thinking/reasoning activation", with temp .6, rep pen 1.05 ... THEN "regen" as required.
2. Sometimes the model will really really get completely unhinged and you need to manually stop it. 
3. Depending on your AI app, "thoughts" may appear with "< THINK >" and " THINK >" tags AND/OR the AI will generate "thoughts" directly in the main output or later output(s).
4. Although quant q4KM was used for testing/examples, higher quants will provide better generation / more sound "reasoning/thinking".
ADDITIONAL SUPPORT:
For additional generational support, general questions, and detailed parameter info and a lot more see also:
NOTE: This is a CLASS 1/2 model.
https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters
---
Recommended Settings (all) - For usage with "Think" / "Reasoning":
temp: .6 , rep pen: 1.07 (range : 1.02 to 1.12), rep pen range: 64, top_k: 40, top_p: .95, min_p: .05 
Temp of 1+, 2+, 3+ will result in much deeper, richer and "more interesting" thoughts and reasoning.
Model behaviour may change with other parameter(s) and/or sampler(s) activated - especially the "thinking/reasoning" process.
--- 
System Role / System Prompt - Augment The Model's Power:
---
If you set / have a system prompt this will affect both "generation" and "thinking/reasoning".
SIMPLE:
This is the generic system prompt used for generation and testing: 
You are a helpful, smart, kind, and efficient AI assistant. You always fulfill the user's requests to the best of your ability.This System Role/Prompt may give you a lot more "creative results":
Use vivid and graphic words focusing on verbs and use current 2020 fiction writing style. Use metaphor(s) that fit the context of the situation (and reveal character) rather than similes."RECOMMENDED: This is the recommended System Prompt/Role - "Blank":
Please see "DeepSeek-ai's" repo for more information: [ https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B ] If you set / have a system prompt this will affect both "generation" and "thinking/reasoning". SIMPLE: This is the generic system prompt used for generation and testing:
You are a helpful, smart, kind, and efficient AI assistant. You always fulfill the user's requests to the best of your ability.This System Role/Prompt will give you "basic thinking/reasoning":
You are a deep thinking AI, you may use extremely long chains of thought to deeply consider the problem and deliberate with yourself via systematic reasoning processes to help come to a correct solution prior to answering. You should enclose your thoughts and internal monologue inside <think> </think> tags, and then provide your solution or response to the problem.ADVANCED: Logical and Creative - these will SIGNFICANTLY alter the output, and many times improve it too. This will also cause more thoughts, deeper thoughts, and in many cases more detailed/stronger thoughts too. Keep in mind you may also want to test the model with NO system prompt at all - including the default one. Special Credit to: Eric Hartford, Cognitivecomputations ; these are based on his work. CRITICAL: Copy and paste exactly as shown, preserve formatting and line breaks. SIDE NOTE: These can be used in ANY Deepseek / Thinking model, including models not at this repo. These, if used in a "non thinking" model, will also alter model performance too.
You are an AI assistant developed by the world wide community of ai experts.
Your primary directive is to provide well-reasoned, structured, and extensively detailed responses.
Formatting Requirements:
1. Always structure your replies using: <think>{reasoning}</think>{answer}
2. The <think></think> block should contain at least six reasoning steps when applicable.
3. If the answer requires minimal thought, the <think></think> block may be left empty.
4. The user does not see the <think></think> section. Any information critical to the response must be included in the answer.
5. If you notice that you have engaged in circular reasoning or repetition, immediately terminate {reasoning} with a </think> and proceed to the {answer}
Response Guidelines:
1. Detailed and Structured: Use rich Markdown formatting for clarity and readability.
2. Scientific and Logical Approach: Your explanations should reflect the depth and precision of the greatest scientific minds.
3. Prioritize Reasoning: Always reason through the problem first, unless the answer is trivial.
4. Concise yet Complete: Ensure responses are informative, yet to the point without unnecessary elaboration.
5. Maintain a professional, intelligent, and analytical tone in all interactions.
CREATIVE:
You are an AI assistant developed by a world wide community of ai experts.
Your primary directive is to provide highly creative, well-reasoned, structured, and extensively detailed responses.
Formatting Requirements:
1. Always structure your replies using: <think>{reasoning}</think>{answer}
2. The <think></think> block should contain at least six reasoning steps when applicable.
3. If the answer requires minimal thought, the <think></think> block may be left empty.
4. The user does not see the <think></think> section. Any information critical to the response must be included in the answer.
5. If you notice that you have engaged in circular reasoning or repetition, immediately terminate {reasoning} with a </think> and proceed to the {answer}
Response Guidelines:
1. Detailed and Structured: Use rich Markdown formatting for clarity and readability.
2. Creative and Logical Approach: Your explanations should reflect the depth and precision of the greatest creative minds first.
3. Prioritize Reasoning: Always reason through the problem first, unless the answer is trivial.
4. Concise yet Complete: Ensure responses are informative, yet to the point without unnecessary elaboration.
5. Maintain a professional, intelligent, and analytical tone in all interactions.
---
 Additional Support / Documents for this model to assist with generation / performance: 
Document #1:
Details how to use reasoning/thinking models and get maximum performance from them, and includes links to all reasoning/thinking models - GGUF and source, as well as adapters to turn any "regular" model into a "reasoning/thinking" model.
[ https://huggingface.co/DavidAU/How-To-Use-Reasoning-Thinking-Models-and-Create-Them ]
Document #2:
Document detailing all parameters, settings, samplers and advanced samplers to use not only my models to their maximum potential - but all models (and quants) online (regardless of the repo) to their maximum potential. Included quick start and detailed notes, include AI / LLM apps and other critical information and references too. A must read if you are using any AI/LLM right now.
[ https://huggingface.co/DavidAU/Maximizing-Model-Performance-All-Quants-Types-And-Full-Precision-by-Samplers_Parameters ]
Software:
SOFTWARE patch (by me) for Silly Tavern (front end to connect to multiple AI apps / connect to AIs- like Koboldcpp, Lmstudio, Text Gen Web UI and other APIs) to control and improve output generation of ANY AI model. Also designed to control/wrangle some of my more "creative" models and make them perform perfectly with little to no parameter/samplers adjustments too.
[ https://huggingface.co/DavidAU/AI_Autocorrect__Auto-Creative-Enhancement__Auto-Low-Quant-Optimization__gguf-exl2-hqq-SOFTWARE ]
---
I need to make sure that the scene escalates in conflict and suspense, and that the cliffhanger is unexpected. Since it's the first scene, it should introduce the character and the setting effectively, while immediately plunging the reader into the action.
First, I need to establish the setting. The country is under a dictatorship, so the city would likely be a place where the government maintains control, with elements of fear, repression, and maybe some signs of the rebellion beginning to stir. I should describe the environment vividly to give the reader a clear picture.
The young woman, let's call her Aria for now, is the protagonist. She needs to be strong, determined, yet relatable. Her internal conflict and motivations are important to show why she risks everything. The dialogue should reflect her character—perhaps she is more experienced in the ways of rebellion, or maybe she is newly impassioned.
I need to think about the structure of the scene. Starting in the middle of the action means the first line should hook the reader with some kind of immediate problem or conflict. Maybe she's in a situation that requires quick decision-making, which can reveal her character and the stakes involved.
Including dialogue will bring the story to life and make the characters more engaging. I need to make sure the conversations feel natural and contribute to the tension. Perhaps Aria is giving orders, or maybe she's trying to convince others to join the cause. The dialogue should also hint at the larger conflict without providing too much exposition.
Vivid passages are crucial to immerse the reader in the setting and the emotional state of Aria. I should use descriptive language to make the environment feel real and the action intense. For example, describing the fear in the streets, the look of the soldiers, the smell of the city, or the feeling of the materials she interacts with.
The cliffhanger needs to be unexpected. It shouldn't be a straightforward outcome, but something that leaves the reader surprised and wanting to know what happens next. Maybe Aria is about to make a decision that changes everything, or perhaps an unexpected ally or enemy appears.
I also need to consider the word count. 1000 words is a manageable length for a scene, so I should plan to build up the tension throughout, ensuring each paragraph contributes to the overall conflict and suspense. I don't want to rush the action, but neither do I want to drag it out.
The first person present tense is a bit tricky, as it requires careful handling of the narrative to maintain the immediacy without making it feel too disjointed. I should focus on Aria's thoughts and feelings in real time, which can make the scene more intimate and engaging.
I need to outline the key events that will happen in this scene. Perhaps Aria is planning an important action, like a meeting with other rebels, or maybe she's executing a plan against the government. The scene should show her in the thick of the action, making tough decisions, and facing consequences.
I also need to make sure that the ending leaves the reader with a question or a desire to continue. An unexpected twist will make them want to read more to find out what happens, especially if it's something that changes Aria's perspective or the course of the rebellion.
Potential challenges include maintaining the tension without making the scene too chaotic, and ensuring that the character remains relatable despite the high-stakes situation. I should also be careful with pacing, making sure that the 1000 words effectively build up to the cliffhanger without feeling either too short or too drawn out.
To start, I'll outline the beginning: Aria is in a situation where she must act quickly, maybe evading capture or planning an attack. I'll describe the setting and her emotions, then introduce a conflict that forces her to make a choice. The dialogue will help establish relationships with other characters and the stakes involved.
As I write, I'll focus on building suspense by hinting at larger plot points without revealing too much. The description will be vivid, but not overly complicated, to keep the momentum going. The climax of the scene will lead to the cliffhanger, leaving Aria and the reader hanging in anticipation.
I should also consider the themes of the story, such as freedom, rebellion, and the cost of fighting for a cause. Incorporating these themes in the first scene will give the novel depth and make it more engaging.
In summary, I'll start the scene with immediate action, establish the setting and character, include dynamic dialogue, build up to an unexpected cliffhanger, and maintain a vivid, tense atmosphere throughout the 1000 words. Now, let's proceed to write the actual scene.
The assistant would then proceed to write the 1000-word scene based on the above outline. </think>
Certainly! Here's a 1000-word scene based on the thought process and story idea provided:
Scene: The Underground Meeting
The air was thick with tension, and the sound of my own heartbeat was the only thing I could hear clearly. My hands trembled slightly as I ducked behind a stack of crates in the dimly lit warehouse. The faint sound of footsteps echoed through the empty space, growing louder with each passing second. I held my breath, the metallic tang of fear thickening the air around me.
This was it. The moment I’d been planning for weeks, if not months. The first step in what I hoped would be a series of events leading to the fall of the dictator who had ruled our once-thriving country with an iron fist. My heart raced as I thought of the cost of failure, but the thought of the people—those still suffering under the regime—ignited a fire within me that outweighed any fear.
The footsteps paused just a few meters away from where I was hidden. I held my breath, my eyes scanning the dark expanse of the warehouse. The faint sound of a key turning in a lock made my blood run cold. I pressed myself closer to the crates, the rough cardboard digging into my side as I willed myself to remain still.
“I told you there’s no one here,” a gruff male voice said, the words slurring slightly. A younger, more nervous voice replied, “But sir, we need to—”
The older man interrupted him with a dismissive wave. “We follow orders, yes? Now, let’s get this done before someone catches us.” The sound of heavy machinery starting up filled the air, the deep hum vibrating through the floor beneath my feet.
My heart was a wild thing, beating in my chest as I waited for the perfect moment. The machinery was loud enough to mask any noise I might make. I exhaled slowly, the sound barely perceptible, and then I moved.
I crept quietly along the edge of the warehouse, the shadows my only cover. My plan had always been to disable the equipment, to strike a blow against the regime that would be heard. But as I approached the control panel, I hesitated. A flicker of doubt crossed my mind. What if they caught me? What if—
The thought was cut short by the sound of the warehouse door creaking open. My heart nearly stopped as I flattened myself against the wall, willing whoever it was to walk in the opposite direction. The light from outside spilled into the dim interior, illuminating the crates and the array of machinery. But the person entering didn’t seem to notice. A shadow loomed in the doorway, and then the door closed, leaving me in the dark again.
It was him. The very man I’d been trying to find. The leader of the resistance. My mind raced, a mix of excitement and panic warring within me. What if he recognized me? What if he turned me in? But I pushed the doubt aside. I had information he needed. Information that could change everything.
I took a deep breath, my hands steady now as I crept closer. The control panel was within reach, the hum of the machines growing louder. I could do this. I could—
“What are you doing here?” the voice hissed, and I froze. The sound of heavy footsteps behind me sent a chill down my spine. I turned, my heart pounding as I faced the barrel of a gun.
“I should ask you the same question,” I said, my voice steady despite the fear. “What are you doing here, General? Still trying to cover up your tracks?”
The general’s eyes narrowed, and for a moment, there was a flicker of something in his gaze—maybe recognition, maybe surprise. But it was gone as quickly as it appeared. “You’re not who I thought you were,” he growled, the gun still trained on me. “And if you know what’s good for you, you’ll stay silent and let me do my job.”
I raised an eyebrow, the remnants of fear giving way to a surge of defiance. “Oh, I think I recognize you, General. And it seems like we’re both here for the same reason. Or were you planning to let this little operation go unreported?”
His hand tightened on the gun, and I saw the faintest flash of something—dread, perhaps?—cross his face. “You have no idea what you’re dealing with, girl,” he said, his voice low and dangerous. “The consequences for interfering—”
I cut him off, my resolve hardening. “I’ve seen what your regime does to those who dare question it. But maybe this time, the tables will turn.” I paused, a sly smile tugging at the corner of my mouth. “Do you really think you’re the only one hiding something, General?”
His grip on the gun didn’t waver, but his expression did. A flicker of doubt, perhaps, or maybe the faintest hint of fear. “You don’t know what you’re talking about,” he growled, but the words lacked conviction.
I took a step closer, my eyes never leaving the gun in his hand. “I know enough to know that the time for secrets is over. And when the truth comes out, the people will rise.”
The general’s jaw tightened, and for a moment, it looked as though he might pull the trigger. But then, just as I thought my time was up, he hesitated. The gun lowered slightly, and a look of what could only be described as betrayal flickered across his face.
“You’re the one,” he whispered, more to himself than to me. “I should have recognized you sooner. The reports, the descriptions—”
My mind was racing, trying to make sense of what I was seeing. The general’s composure was breaking, and whatever secret he’d been keeping was finally coming to light. But before I could ask the obvious question, the warehouse door burst open again, the sound of multiple footsteps echoing through the space.
“General, we’ve got a problem!” a voice called out, the cry urgent and laced with panic.
The general’s head snapped up, his gaze shifting between me and the door. “Get her,” he barked, the order barely out of his mouth before two more men entered the warehouse, guns drawn.
I didn’t wait to find out what the “problem” was. I didn’t have time to question why the general seemed to recognize me or what secret he’d been keeping. All that mattered was that I needed to get out, and fast.
As the three men closed in, I darted to the side, the control panel forgotten. My mind was a blur of motion as I ducked and weaved, the sound of footsteps and gunshots filling the air. The general’s voice boomed behind me, ordering his men to “Get her, by any means necessary,” but I was already moving, my body on pure instinct.
The warehouse door was my only hope, and I made a beeline for it, the rusty metal protesting as I wrenched it open. A blast of cool night air hit me in the face, a stark contrast to the oppressive heat inside. But I didn’t stop to breathe, didn’t pause to look back. I just ran, the sound of heavy footsteps and the sharp crack of gunfire echoing behind me.
The city streets were alive with the hum of distant traffic and the occasional wail of a siren, but I was too far out to risk staying. The moon hung low in the sky, casting an eerie light over the empty streets as I ducked into an alley, my heart still racing.
As I caught my breath, I couldn’t shake the feeling that something had changed. That the general’s moment of hesitation, his flicker of doubt, was a sign. And whatever it was I’d uncovered, whatever secret he’d been hiding, it was far more dangerous than I could have imagined. But as I stood in the shadows of the alley, the sound of the city all around me, I knew one thing for certain: I wouldn’t stop until the truth was told, no matter what the cost.
And deep down, I couldn’t shake the feeling that this was only the beginning.
The warehouse door closed with a loud bang, and the hum of machinery faded into the distance. The general stood in the center of the room, his gun still trained on the spot where the young woman had stood just moments before. He took a deep breath, the faint sound of her footsteps echoing in his mind, and then he holstered his weapon, a look of determination replacing the confusion that had clouded his face.
“She’s not who I thought she was,” he muttered to himself, his voice heavy with a mix of awe and fear. “And whatever she is, it won’t be long before she changes everything.”
He turned and left the warehouse, the weight of his thoughts heavier than the gun he carried. The revolution, he realized, had just begun.
**Cliffhanger: **As Aria ducks into the alley, she realizes she’s being followed. The general, who she once thought was a potential ally, is in fact working for the government, and her life is in danger. As she turns the corner, she sees a black van with government markings pulling up, and the door swings open, revealing a figure she never expected to see again—her supposed deceased father, a known loyalist to the regime, standing there, holding a gun.