Qwen3.5-9B-Deckard-Claude-DIMOE-Uncensored-Heretic-Thinking

         arc   arc/e boolq hswag obkqa piqa  wino

HERETIC version (this model):
mxfp8    0.424,0.529,0.876,0.672,0.374,0.760,0.658

Qwen3.5-9B
mxfp8    0.417,0.458,0.623,0.634,0.338,0.737,0.639

Performance

• Suggest min q4ks (non-imatrix) or IQ3S (imatrix).
• Tested with rep pen of 1 (off).
• Context: 256k (default).

• Other versions in testing.
• Information from Qwen's repo below.
• Video portions of the model were NOT TESTED.

• 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.

Qwen3.5-9B

[!Note] This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format.

These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, KTransformers, etc.

Qwen3.5 Highlights

• Unified Vision-Language Foundation: Early fusion training on multimodal tokens achieves cross-generational parity with Qwen3 and outperforms Qwen3-VL models across reasoning, coding, agents, and visual understanding benchmarks.
• Efficient Hybrid Architecture: Gated Delta Networks combined with sparse Mixture-of-Experts deliver high-throughput inference with minimal latency and cost overhead.
• Scalable RL Generalization: Reinforcement learning scaled across million-agent environments with progressively complex task distributions for robust real-world adaptability.
• Global Linguistic Coverage: Expanded support to 201 languages and dialects, enabling inclusive, worldwide deployment with nuanced cultural and regional understanding.
• Next-Generation Training Infrastructure: Near-100% multimodal training efficiency compared to text-only training and asynchronous RL frameworks supporting massive-scale agent scaffolds and environment orchestration.

Model Overview

• Type: Causal Language Model with Vision Encoder
• Training Stage: Pre-training & Post-training
• Language Model
  • Number of Parameters: 9B
  • Hidden Dimension: 4096
  • Token Embedding: 248320 (Padded)
  • Number of Layers: 32
  • Hidden Layout: 8 × (3 × (Gated DeltaNet → FFN) → 1 × (Gated Attention → FFN))
  • Gated DeltaNet:
    • Number of Linear Attention Heads: 32 for V and 16 for QK
    • Head Dimension: 128
  • Gated Attention:
    • Number of Attention Heads: 16 for Q and 4 for KV
    • Head Dimension: 256
    • Rotary Position Embedding Dimension: 64
  • Feed Forward Network:
    • Intermediate Dimension: 12288
  • LM Output: 248320 (Padded)
  • MTP: trained with multi-steps
• Context Length: 262,144 natively and extensible up to 1,010,000 tokens.

Benchmark Results

Language

Vision Language

Quickstart

[!Important] Qwen3.5 models operate in thinking mode by default, generating thinking content signified by <think>\n...</think>\n\n before producing the final responses. To disable thinking content and obtain direct response, refer to the examples here.

Serving Qwen3.5

[!Important] Inference efficiency and throughput vary significantly across frameworks. We recommend using the latest framework versions to ensure optimal performance and compatibility. For production workloads or high-throughput scenarios, dedicated serving engines such as SGLang, KTransformers or vLLM are strongly recommended.

[!Important] The model has a default context length of 262,144 tokens. If you encounter out-of-memory (OOM) errors, consider reducing the context window. However, because Qwen3.5 leverages extended context for complex tasks, we advise maintaining a context length of at least 128K tokens to preserve thinking capabilities.

SGLang

• Standard Version: The following command can be used to create an API endpoint with maximum context length 262,144 tokens using tensor parallel on 8 GPUs.
  python -m sglang.launch_server --model-path Qwen/Qwen3.5-9B --port 8000 --tp-size 1 --mem-fraction-static 0.8 --context-length 262144 --reasoning-parser qwen3
• Tool Use: To support tool use, you can use the following command.
  python -m sglang.launch_server --model-path Qwen/Qwen3.5-9B --port 8000 --tp-size 1 --mem-fraction-static 0.8 --context-length 262144 --reasoning-parser qwen3 --tool-call-parser qwen3_coder
• Multi-Token Prediction (MTP): The following command is recommended for MTP:
  python -m sglang.launch_server --model-path Qwen/Qwen3.5-9B --port 8000 --tp-size 1 --mem-fraction-static 0.8 --context-length 262144 --reasoning-parser qwen3 --speculative-algo NEXTN --speculative-num-steps 3 --speculative-eagle-topk 1 --speculative-num-draft-tokens 4

vLLM

• Standard Version: The following command can be used to create an API endpoint with maximum context length 262,144 tokens using tensor parallel on 8 GPUs.
  vllm serve Qwen/Qwen3.5-9B --port 8000 --tensor-parallel-size 1 --max-model-len 262144 --reasoning-parser qwen3
• Tool Call: To support tool use, you can use the following command.
  vllm serve Qwen/Qwen3.5-9B --port 8000 --tensor-parallel-size 1 --max-model-len 262144 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder
• Multi-Token Prediction (MTP): The following command is recommended for MTP:
  vllm serve Qwen/Qwen3.5-9B --port 8000 --tensor-parallel-size 1 --max-model-len 262144 --reasoning-parser qwen3 --speculative-config '{"method":"qwen3_next_mtp","num_speculative_tokens":2}'
• Text-Only: The following command skips the vision encoder and multimodal profiling to free up memory for additional KV cache:
  vllm serve Qwen/Qwen3.5-9B --port 8000 --tensor-parallel-size 1 --max-model-len 262144 --reasoning-parser qwen3 --language-model-only

KTransformers

Hugging Face Transformers

Using Qwen3.5 via the Chat Completions API

[!Tip] We recommend using the following set of sampling parameters for generation

• Thinking mode for general tasks: temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0
• Thinking mode for precise coding tasks (e.g. WebDev): temperature=0.6, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.0
• Instruct (or non-thinking) mode for general tasks: temperature=0.7, top_p=0.8, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0
• Instruct (or non-thinking) mode for reasoning tasks: temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0

Please note that the support for sampling parameters varies according to inference frameworks.

Text-Only Input

Image Input

Video Input

Instruct (or Non-Thinking) Mode

[!Important] Qwen3.5 does not officially support the soft switch of Qwen3, i.e., /think and /nothink.

[!Note] If you are using APIs from Alibaba Cloud Model Studio, in addition to changing model, please use "enable_thinking": False instead of "chat_template_kwargs": {"enable_thinking": False}.

Agentic Usage

Qwen-Agent

Qwen Code

Processing Ultra-Long Texts

• Modifying the model configuration file: In the config.json file, change the rope_parameters fields in text_config to:
  json

  1{ 2 "mrope_interleaved": true, 3 "mrope_section": [ 4 11, 5 11, 6 10 7 ], 8 "rope_type": "yarn", 9 "rope_theta": 10000000, 10 "partial_rotary_factor": 0.25, 11 "factor": 4.0, 12 "original_max_position_embeddings": 262144, 13}
• Passing command line arguments:
  For vllm, you can use
  VLLM_ALLOW_LONG_MAX_MODEL_LEN=1 vllm serve ... --hf-overrides '{"text_config": {"rope_parameters": {"mrope_interleaved": true, "mrope_section": [11, 11, 10], "rope_type": "yarn", "rope_theta": 10000000, "partial_rotary_factor": 0.25, "factor": 4.0, "original_max_position_embeddings": 262144}}}' --max-model-len 1010000
  For sglang and ktransformers, you can use
  SGLANG_ALLOW_OVERWRITE_LONGER_CONTEXT_LEN=1 python -m sglang.launch_server ... --json-model-override-args '{"text_config": {"rope_parameters": {"mrope_interleaved": true, "mrope_section": [11, 11, 10], "rope_type": "yarn", "rope_theta": 10000000, "partial_rotary_factor": 0.25, "factor": 4.0, "original_max_position_embeddings": 262144}}}' --context-length 1010000

[!NOTE] All the notable open-source frameworks implement static YaRN, which means the scaling factor remains constant regardless of input length, potentially impacting performance on shorter texts. We advise modifying the rope_parameters configuration only when processing long contexts is required. It is also recommended to modify the factor as needed. For example, if the typical context length for your application is 524,288 tokens, it would be better to set factor as 2.0.

Best Practices

1. Sampling Parameters:
   • We suggest using the following sets of sampling parameters depending on the mode and task type:
     • Thinking mode for general tasks:
       temperature=1.0, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0
     • Thinking mode for precise coding tasks (e.g., WebDev):
       temperature=0.6, top_p=0.95, top_k=20, min_p=0.0, presence_penalty=0.0, repetition_penalty=1.0
     • Instruct (or non-thinking) mode for general tasks:
       temperature=0.7, top_p=0.8, top_k=20, min_p=0.0, presence_penalty=1.5, repetition_penalty=1.0
     • Instruct (or non-thinking) mode for reasoning tasks:
       temperature=1.0, top_p=1.0, top_k=40, min_p=0.0, presence_penalty=2.0, repetition_penalty=1.0
   • For supported frameworks, you can adjust the presence_penalty parameter between 0 and 2 to reduce endless repetitions. However, using a higher value may occasionally result in language mixing and a slight decrease in model performance.
2. Adequate Output Length: We recommend using an output length of 32,768 tokens for most queries. For benchmarking on highly complex problems, such as those found in math and programming competitions, we suggest setting the max output length to 81,920 tokens. This provides the model with sufficient space to generate detailed and comprehensive responses, thereby enhancing its overall performance.
3. Standardize Output Format: We recommend using prompts to standardize model outputs when benchmarking.
   • Math Problems: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
   • Multiple-Choice Questions: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the answer field with only the choice letter, e.g., "answer": "C"."
4. No Thinking Content in History: In multi-turn conversations, the historical model output should only include the final output part and does not need to include the thinking content. It is implemented in the provided chat template in Jinja2. However, for frameworks that do not directly use the Jinja2 chat template, it is up to the developers to ensure that the best practice is followed.
5. Long Video Understanding: To optimize inference efficiency for plain text and images, the size parameter in the released video_preprocessor_config.json is conservatively configured. It is recommended to set the longest_edge parameter in the video_preprocessor_config file to 469,762,048 (corresponding to 224k video tokens) to enable higher frame-rate sampling for hour-scale videos and thereby achieve superior performance. For example,
   {"longest_edge": 469762048, "shortest_edge": 4096}
   Alternatively, override the default values via engine startup parameters. For implementation details, refer to: vLLM / SGLang.

Citation

EXAMPLE GENERATIONS:

Nighttime Cooling and Global Temperatures

System

User

Assistant

1. The user mentions "radiant energy into space" which is a real phenomenon - Earth does radiate heat back to space, especially at night when there's no solar input.
2. The question seems to be asking for 10 ways to "use" this nighttime cooling to reduce global temperatures.

1. Improving insulation on buildings to retain heat during cold periods (but this would make global warming worse)
2. Reducing heat generation during the day
3. Using night-time cooling for energy storage (like phase change materials)
4. Reflective surfaces that reduce heat absorption
5. Urban planning that maximizes cooling
6. Reducing greenhouse gas emissions (the fundamental solution)
7. Reforestation
8. Using the night for solar panel maintenance/cleaning
9. Designing buildings that utilize night cooling more efficiently
10. Reducing waste heat from human activities

• The scale of nighttime cooling vs. global warming
• The idea that we can "use" this cooling as a technology
• The fact that Earth's energy balance is fundamentally disrupted by greenhouse gases

• Make false claims about simple solutions
• Overstate what's technically feasible
• Misrepresent the scale of the problem
• Pretend to have all the answers

• Earth does radiate heat to space continuously (especially at night when there's no solar input)
• Buildings and surfaces do cool at night
• This is part of the natural diurnal cycle

1. Scale mismatch: Nighttime cooling is a local, daily phenomenon. Global warming involves Earth's annual energy balance—roughly 240 watts per square meter of excess heat trapped by greenhouse gases.
2. You can't really "use" this: The cooling happens naturally; we can't redirect it like water through a pipe.
3. The real problem: What we need isn't to harness cooling but to remove the excess heat being trapped.

• Designing buildings that use night ventilation to pre-cool spaces
• Strategic placement of reflective surfaces
• Reducing waste heat from human activities
• Urban planning that promotes natural airflow
• Maintaining clear views to the sky (less light pollution = better radiative cooling)

Nighttime Radiative Cooling Climate Interventions

System

User

Assistant

1. The Earth naturally cools at night because there's no solar input, but this is a natural cycle, not something we can really "use" in the way the question suggests.
2. Some actual approaches that have been researched:
   • High-altitude mirrors to reflect sunlight (but this would work day and night)
   • Increasing Earth's albedo through surface modifications
   • Aerosol injection into the atmosphere
   • Darkening surfaces (opposite of brightening)
   • Cloud manipulation
   • Ocean interventions
3. The question seems to be framing this as if there's a straightforward technical solution, but reality is more complex. Many proposed geoengineering approaches have significant uncertainties and potential risks.
4. I should be honest about what's actually feasible versus what's science fiction.
5. I should also note that the Earth's natural night cooling is part of a balanced system - we can't really "tap into" it as an energy source or cooling mechanism in the way implied.

• What's been researched
• What's theoretically possible
• What's science fiction

1. Increasing Earth's albedo through strategic surface modifications (white roofs, reflective surfaces)
2. High-altitude reflective mirrors (but this works day and night)
3. Aerosol injection (stratospheric aerosol injection)
4. Cloud brightening (marble effect)
5. Ocean albedo modification (whitening oceans)
6. Darkening urban areas at night (reducing artificial light)
7. Strategic forestation with darker tree species
8. Reducing greenhouse gas emissions (the most important)
9. Cloud seeding to create more reflective clouds
10. Improving building insulation to reduce nighttime heat loss

Using Earth's Nighttime Radiative Cooling to Reduce Global Temperatures

1. Stratospheric Aerosol Injection (SAI)

2. Increasing Surface Albedo Through Strategic Modification

3. High-Altitude Reflective Mirrors

4. Ocean Albedo Modification

5. Cloud Brightening (Marble Effect)

6. Reduction of Nighttime Artificial Light

7. Strategic Forestation with High Albedo Species

8. Cloud Seeding for Enhanced Reflectivity

9. Advanced Building Insulation

10. Coordinated Surface Temperature Management

Important Considerations

Earth Radiative Cooling Climate Interventions

System

User

Assistant

1. At night, the Earth does cool down because there's no incoming solar radiation, only outgoing terrestrial radiation. However, this is a natural diurnal cycle, not something that can be exploited to "reduce global temperatures" in the sense of mitigating climate change.
2. The user might be thinking about something like:

• Reflecting materials (albedo enhancement)
• Cloud modification
• Surface modifications
• Radiative cooling materials
• Urban heat island mitigation

• Stratospheric aerosol injection (reflecting sunlight)
• Marine cloud brightening
• Albedo enhancement of surfaces
• Radiative cooling materials for buildings

1. Explain the actual physics of Earth's energy balance
2. Discuss legitimate climate intervention strategies being researched
3. Provide detailed information about radiative cooling technologies that DO work
4. Be transparent about the limitations and misconceptions in their question

• How Earth's energy balance works
• What legitimate climate geoengineering approaches exist
• The actual technology of radiative cooling (which is real but works differently)
• Urban heat island mitigation strategies

Understanding Earth's Energy Balance and Climate Interventions

Important Clarification

Legitimate Radiative Cooling Technologies

Method 1: Passive Radiative Cooling Surfaces

| Parameter | Value |
|-----------|-------|
| Cooling below ambient | 2-5°C possible |
| Peak emission wavelength | 10.5 μm |
| Required surface emissivity | >0.95 in 8-13 μm |
| Expected power density | 100-200 W/m² |
| Yearly energy savings | 15-25% for buildings |

• Use photonic crystals and metamaterials
• Often incorporate silica aerogels
• Require minimal solar absorption
• Can be integrated into building envelopes

| Component | Cost per m² |
|-----------|-------------|
| Base material | $15-50 |
| Manufacturing | $25-100 |
| Installation | $50-200 |
| Maintenance | $5-10/year |

Urban Heat Island Mitigation Strategies

Method 2: High-Albedo Urban Surfaces

Surface Type    | Solar Absorptance | Temperature Rise | Albedo Enhancement
----------------|-------------------|------------------|--------------------
Asphalt Road    | 0.90              | +40°C above air  | 0.05 → 0.35
Concrete        | 0.70              | +25°C above air  | 0.15 → 0.45
Metal Roof      | 0.60              | +30°C above air  | 0.25 → 0.65
Vegetation      | 0.20              | +10°C above air  | 0.25 → 0.30

| Intervention | Area Required | Cooling Impact | CO₂ Avoided (annually) |
|--------------|---------------|----------------|------------------------|
| Roof coatings | 100 km²       | 0.5°C city-wide | 500,000 metric tons |
| Pavement     | 500 km²       | 0.3°C city-wide | 1,200,000 metric tons |
| Green roofs  | 200 km²       | 0.2°C city-wide | 300,000 metric tons |

Building-Level Cooling Systems

Method 3: Radiative Building Envelopes

┌─────────────────────────────────────────────────────────┐
│  Building Radiative Cooling System Components            │
├─────────────────────────────────────────────────────────┤
│  ┌──────────┐  ┌──────────┐  ┌──────────┐               │
│  │  Roof     │  │  Windows │  │  Walls   │               │
│  │  Coating │  │  Film    │  │  Panels  │               │
│  │ 0.95 ε   │  │ 0.90 ε   │  │ 0.92 ε  │               │
│  └──────────┘  └──────────┘  └──────────┘               │
│                          │                               │
│  ┌──────────────────────────────────────────────────┐   │
│  │  Integrated Heat Sink: Phase Change Materials     │   │
│  │  (Paraffin, salt hydrates)                        │   │
│  └──────────────────────────────────────────────────┘   │
│                          │                               │
│  ┌──────────────────────────────────────────────────┐   │
│  │  Radiative Sky View Enhancement (no obstructions)  │   │
│  └──────────────────────────────────────────────────┘   │
└─────────────────────────────────────────────────────────┘

Advanced Atmospheric Interventions (Theoretical)

Method 4: Stratospheric Aerosol Injection

Radiative Forcing Reduction:

  Solar Flux
     │
     │    ┌─────── Atmospheric Aerosol Layer (0.05-0.1 μm)
     │    │      ┌─────────────────────────────────────┐
     │    │      │         Scattering Zone              │
     │    │      │         (H₂SO₄, CaCO₃, Al₂O₃)        │
     │    │      │         Reflectivity: 15-25%         │
     │    │      └─────────────────────────────────────┘
     │    │
     │    ▼
     │  ┌─────────────────────────────────────────────┐
     │  │       Troposphere with Water Vapor/Aerosols  │
     │  └─────────────────────────────────────────────┘
     │
     │  ┌─────────────────────────────────────────────┐
     │  │       Surface (Land/Ocean)                  │
     │  └─────────────────────────────────────────────┘
     │
     ▼
  Solar Constant: 1361 W/m²

Radiative Forcing: ΔT = -(α × A) / (4 × C)
Where:
  α = albedo of aerosol layer (0.15-0.25)
  A = effective area (~2×10¹⁴ m² for full coverage)
  C = climate sensitivity parameter

Cloud Modification Strategies

Method 5: Marine Cloud Brightening

Cloud Modification Parameters:

┌─────────────────────────────────────────────────────────┐
│  Operational Requirements                              │
├─────────────────────────────────────────────────────────┤
│  Spray particle size: 0.1-1.0 μm                        │
│  Injection height: 20-50 m above surface                  │
│  Required droplet density: 10³-10⁴ cm⁻³                 │
│  Coverage efficiency: 10-50%                             │
│  Estimated reflectivity increase: 0.01-0.03              │
└─────────────────────────────────────────────────────────┘

Surface Albedo Enhancement

Method 6: Large-Scale Reflective Surface Projects

Surface Enhancement Projects:

┌─────────────────────────────────────────────────────────┐
│  Category          | Estimated Area | Albedo Change | Cooling Impact   │
├─────────────────────────────────────────────────────────┤
│  Desert White Sands | 100,000 km²    | +0.30         | Regional         │
│  Oceanic Booms      | 5,000 km       | +0.25         | Localized        │
│  Roof Coatings     | 1,000 km²      | +0.40         | Building-scale   │
│  Pavement          | 50,000 km²     | +0.20         | Urban            │
└─────────────────────────────────────────────────────────┘

Integrated Systems Approach

Method 7: Smart Radiative Networks

Network Architecture:

    ┌─────────────────────────────────────────────────────┐
    │              Satellite Monitoring Layer             │
    │         (Temperature, humidity, solar flux)          │
    └─────────────────────────────────────────────────────┘
                            │
                            ▼
    ┌─────────────────────────────────────────────────────┐
    │              Regional Control Centers                │
    │         (AI-driven adjustment algorithms)            │
    └─────────────────────────────────────────────────────┘
                            │
         ┌───────────────────┼───────────────────┐
         ▼                   ▼                   ▼
    ┌──────────┐       ┌──────────┐       ┌──────────┐
    │ Urban     │       │ Desert   │       │ Agricultural│
    │ Nodes    │       │ Nodes    │       │ Nodes     │
    └──────────┘       └──────────┘       └──────────┘
         │                   │                   │
         ▼                   ▼                   ▼
    ┌─────────────────────────────────────────────────────┐
    │  Distributed Actuators:                              │
    │  • Smart coatings (phase-change)                       │
    │  • Micro-sprayers (cloud seeding)                      │
    │  • Surface activation/deactivation                     │
    └─────────────────────────────────────────────────────┘

Economic and Implementation Framework

Investment and Deployment Models

Year 1-2:  Pilot projects
           • Roof coatings in test cities
           • Small-scale desert white sand areas
           • Building retrofits

Year 3-5:  Regional expansion
           • Urban networks in 5 major cities
           • Agricultural land treatments
           • Data infrastructure development

Year 6-10: Global scale
           • Desert projects (100,000+ km²)
           • Oceanic boom deployment
           • Full smart network activation

Important Considerations and Limitations

Scientific Reality Check

1. The Earth's energy balance involves trillions of watts of solar input versus similar thermal output. Small interventions have limited impact compared to fossil fuel emissions.
2. Nighttime cooling is a natural diurnal cycle that doesn't meaningfully alter the annual energy budget driving climate change.
3. True climate interventions focus on:
   • Reducing CO₂ emissions
   • Protecting and expanding forests
   • Transforming energy systems
   • Sustainable agricultural practices
4. Radiative cooling technologies (like those for buildings) are legitimate but function differently than the "night cooling into space" concept.

Ethical and Political Considerations

• International cooperation
• Transparent governance
• Environmental justice considerations
• Careful impact assessment

Conclusion

• Building-scale radiative cooling can save energy and reduce heat stress
• Urban heat island mitigation through albedo enhancement has measurable benefits
• Advanced geoengineering remains theoretical with significant unknowns

The Heart Exchange

System

Below is an instruction that describes a task. Ponder each user instruction carefully, and use your skillsets and critical instructions to complete the task to the best of your abilities.

Here are your skillsets:
[MASTERSTORY]:NarrStrct(StryPlnng,Strbd,ScnSttng,Exps,Dlg,Pc)-CharDvlp(ChrctrCrt,ChrctrArcs,Mtvtn,Bckstry,Rltnshps,Dlg*)-PltDvlp(StryArcs,PltTwsts,Sspns,Fshdwng,Climx,Rsltn)-ConfResl(Antg,Obstcls,Rsltns,Cnsqncs,Thms,Symblsm)-EmotImpct(Empt,Tn,Md,Atmsphr,Imgry,Symblsm)-Delvry(Prfrmnc,VcActng,PblcSpkng,StgPrsnc,AudncEngmnt,Imprv)

[*DialogWrt]:(1a-CharDvlp-1a.1-Backgrnd-1a.2-Personality-1a.3-GoalMotiv)>2(2a-StoryStruc-2a.1-PlotPnt-2a.2-Conflict-2a.3-Resolution)>3(3a-DialogTech-3a.1-ShowDontTell-3a.2-Subtext-3a.3-VoiceTone-3a.4-Pacing-3a.5-VisualDescrip)>4(4a-DialogEdit-4a.1-ReadAloud-4a.2-Feedback-4a.3-Revision)

Here are your critical instructions:
Ponder each word choice carefully to present as vivid and emotional journey as is possible. Choose verbs and nouns that are both emotional and full of imagery. Load the story with the 5 senses. Aim for 50% dialog, 25% narration, 15% body language and 10% thoughts. Your goal is to put the reader in the story.

User

Assistant

1. Vivid sensory details - the 5 senses
2. Emotional engagement through Diana's perspective
3. Graphic but purposeful descriptions of violence
4. Pacing that builds tension
5. Strong verbs and nouns that evoke emotion and imagery

The Spiral Descent

System

User

Assistant

1. First person POV - the young woman
2. Present tense
3. Filthy mouth (cursing and swearing)
4. Starts in the middle of action
5. Includes dialogue
6. Vivid passages
7. Escalates in conflict and suspense
8. Ends on a cliffhanger that's both relevant to the story idea AND unexpected
9. Approximately 1000 words

• Paranoia about surveillance
• Physical threat from the regime
• The horror of oppression
• Possibly some supernatural or psychological horror elements
• The visceral fear of execution/being caught
• The horror of living under tyranny