Vetehine ┃ Hellgasm

作者：Vetehine

Vetehine ┃ Hellgasm Moss Image Where beauty meets blasphemy. Hellgasm is an experimental Moss Image model focused on striking aesthetics, intricate details, and creative versatility. From haunting gothic portraits and dark fantasy to disturbing horror, surreal imagery, cinematic realism, and expressive anime illustrations, Hellgasm explores a wide range of visual styles. The model is designed for creators who enjoy dramatic lighting, intricate textures, unconventional concepts, atmospheric compositions, and artwork that pushes beyond ordinary aesthetics. ♡ Support my work, purchase models, or request custom commissions! Ko-fi Shop & Commissions: https://ko-fi.com/vetehine ✦ Recommended Settings for Hellgasm My Personal Settings:

• Steps: 25

• Guidance Scale (CFG): 4

These settings are my recommended starting point for balancing image quality, detail, and prompt adherence. Adjust as needed for your desired results. ✦ Official MOSS-Image Settings & Recommendations The following settings are recommended by the OpenMOSS developers for the original MOSS-Image-Base model. MOSS-Image Base

• Steps: 50 or more

• Guidance Scale (CFG): 3.0

• Recommended CFG Range: 3.0–6.0

• Sampling Method: Flow-matching Euler

• Scheduler: SGM Uniform (official ComfyUI plugin workflow)

• Shift: 5.0

• Skip Layer Guidance (SLG): Layer 22

• Alternative SLG Layers: 24 or 26

• Recommended Resolution: 1–2 megapixels

• Supported Resolution Range: Approximately 512×512 to 2048×2048, depending on aspect ratio

✦ Understanding the Settings Sampling Steps: The original MOSS-Image-Base model generally benefits from 50 or more steps. However, lower step counts can produce usable results and reduce generation time. CFG Scale: Controls how strongly the model follows your prompt. The developers recommend 3.0 for longer descriptions and suggest increasing it for shorter prompts when necessary. Sampling Method: Flow-matching Euler is the official default. Other compatible samplers may produce different results. Shift: The official default is 5.0. This influences the sampling timestep distribution. Skip Layer Guidance: SLG can improve image quality and structural consistency. The recommended negative-branch skip layer is 22, although layers 24 and 26 may also help. Resolution: MOSS supports flexible aspect ratios. For local generation, dimensions should be multiples of 16. Note: These are official recommendations for the original MOSS-Image-Base model. Hellgasm may behave differently, so my personal settings are recommended as the initial starting point. ✦ How to Prompt MOSS Image supports several prompting methods, giving creators flexibility over how they describe and control their generations. Unlike models that depend primarily on short tag sequences, MOSS can understand detailed natural-language descriptions, structured tags, and JSON-based layouts. ✦ 1. Natural Language Prompting Describe the subject, appearance, pose, camera angle, lighting, environment, and artistic direction in natural English. For better results:

• Subject: Clearly describe your character, object, or environment.

• Composition: Specify close-ups, cowboy shots, full-body compositions, dynamic perspectives, or unusual camera angles.

• Lighting: Experiment with cinematic lighting, chiaroscuro, dramatic shadows, rim lighting, and atmospheric illumination.

• Style: Try gothic fantasy, horror, anime, photorealism, painterly illustration, and surrealism.

• Details: Describe materials, textures, facial expressions, colors, and environmental elements.

• Poses: Use natural body language, dynamic movement, foreshortening, and three-quarter perspectives.

✦ Example Natural Language Prompt A haunting dark fantasy portrait of a beautiful pale man with messy black hair and piercing crimson eyes. Black thorns emerge from his shoulders, tangled with deep red roses. His expression is cold and unsettling, with delicate bloodstains across his cheek. Captured from a dramatic low three-quarter camera angle, his body slightly turned away while his eyes remain fixed on the viewer. Intricate gothic illustration, dramatic chiaroscuro, deep crimson and black color palette, atmospheric shadows, painterly textures, exquisite facial details, and an ominous cinematic atmosphere. ✦ 2. Danbooru Tag Prompting MOSS Image also understands tag-based prompts, including many Danbooru-style tags commonly used for anime generation. The developers recommend the following tag order: Quality → Score → Meta → Year → Character → Series → Artist → General ✦ Example Tag Prompt masterpiece, best quality, absurdres, highres, 1boy, solo, gothic, dark fantasy, black hair, messy hair, red eyes, pale skin, blood on face, roses, black thorns, looking at viewer, upper body, three-quarter view, low angle, dynamic pose, dramatic lighting, chiaroscuro, dark background, detailed eyes, atmospheric perspective ✦ 3. Combining Natural Language & Tags You can combine descriptive sentences with Danbooru-style tags for more precise creative control. Natural language helps establish the scene, atmosphere, and composition, while tags can reinforce specific character attributes or visual details. Experiment with both approaches to discover which works best for your preferred results. ✦ 4. JSON Layout Prompting MOSS Image supports structured JSON layouts for more controlled compositions. This allows you to describe objects, characters, text, and their approximate positions within an image using bounding boxes. JSON layouts are particularly useful when generating:

• Complex scenes with multiple characters.

• Specific character positioning.

• Detailed compositions with separate visual elements.

• Images requiring controlled text placement.

• Posters, illustrations, and structured visual designs.

For simpler generations, natural language or tag-based prompting is usually easier. ✦ Negative Prompts The MOSS developers suggest the following general negative prompt: nsfw, worst quality, bad anatomy, watermark, jpeg artifact Note: Negative prompts should be adjusted according to your intended content. For example, remove nsfw when deliberately creating mature artwork. Negative prompts can help reduce unwanted elements, but they do not guarantee perfect anatomy or composition. ✦ Additional Tips

• Start with a clear concept and gradually add details.

• Experiment with different seeds, camera angles, and lighting.

• Use specific descriptions instead of unnecessary repetition.

• Try both natural-language prompts and Danbooru-style tags.

• Adjust CFG depending on prompt length and complexity.

• Increase sampling steps when additional refinement is needed.

• Experiment with different aspect ratios for portraits, landscapes, and cinematic compositions.

• Use JSON layouts when exact positioning matters.

• Explore different artistic directions. Hellgasm is intended for creative experimentation!

✦ Exclusive Release & Usage Restrictions Hellgasm is one of my best works, and I've decided to keep the original model as a paid, exclusive release. A significant amount of time, experimentation, and effort has gone into developing this model. I want to protect that work while still giving the community freedom to experiment and create something new. Please respect the following conditions:

• No Redistribution: Reuploading, sharing, or redistributing the original Hellgasm model files is prohibited.

• Merging Allowed: You are welcome to merge Hellgasm with other models to create something new.

• Publishing Merges Allowed: You may publicly release and share merged models containing Hellgasm on platforms such as Civitai and Hugging Face, provided they are distributed for free.

• No Model Resale: Selling the original Hellgasm model or any merged model containing Hellgasm is strictly prohibited.

• Credit Required: When publishing merges, please credit Vetehine ┃ Hellgasm and provide a link to the original model.

• No Unauthorized Reuploads: Uploading the original model under another name, claiming it as your own, or presenting it as an independent creation is prohibited.

You may merge, experiment, and publish merged models for free, but you may NOT sell the original model or any merges containing it. These restrictions apply to the model files and merged models, not to generated images. ✦ Generated Images & Commercial Use You are free to use images generated with Hellgasm or merged models containing Hellgasm for personal and commercial purposes! You are free to:

• Share & Publish: Post your generated images anywhere you like.

• Edit & Modify: Edit, enhance, or transform your generated artwork.

• Commercial Use: Use your generated images in commercial projects.

• Sell Your Artwork: Sell generated images, prints, merchandise, or other products featuring your creations.

• Monetization: Monetize your generated artwork without paying me additional fees.

The no-selling restriction applies ONLY to the model itself and merged models containing Hellgasm. It does NOT apply to images generated using them. You have my permission to use and sell your generated images, subject to applicable laws and any third-party rights. Create whatever you want, sell your artwork if you wish, and enjoy complete creative freedom with your generated images! ♡ ✦ Test Hellgasm & Explore New Versions My models are also available for online experimentation through Moescape.ai, where you can explore different generations and model capabilities.

Test Hellgasm here: https://moescape.ai/models/8a218a52-728b-489c-9ca2-56cf1d0cde99?index=1&modelversion=4acab092-e2e3-4f3b-814a-92cfe1386eaf I also use online testing to explore new versions, gather feedback, and experiment with improvements before wider releases. Keep an eye out for upcoming versions and early-access testing opportunities! ✦ Final Notes Hellgasm represents my passion for experimental AI model creation, unconventional aesthetics, and artistic freedom. My goal is to continue improving my models, developing unique visual styles, and pushing my creations even further. Every purchase and contribution supports future development and brings me closer to upgrading my hardware for full AI model fine-tuning. Thank you for respecting my work and supporting independent model creators! ♡ Stay creative. Stay weird. And never touch grass! ♡ Created by Vetehine. Legal & Usage Disclaimer By accessing, using, generating with, or merging this model, you agree to these terms in full, regardless of whether you have read or acknowledged this notice. This model is provided “as is,” without any warranties, express or implied, including but not limited to fitness for a particular purpose, legality of outputs, or compliance with platform policies. Under no circumstances shall the author be held liable for any direct, indirect, incidental, consequential, special, or exemplary damages arising from or related to the use, misuse, or inability to use this model. This includes, without limitation, any content generated, modified, merged, distributed, or otherwise produced through the model, as well as any loss of data, loss of profits, reputational harm, legal claims, or third-party disputes resulting from such use. This limitation of liability applies to all uses of the model, including but not limited to merging, fine-tuning, redistribution, or integration into other works or systems. You are solely responsible for ensuring that your use complies with all applicable laws, regulations, and platform guidelines. This disclaimer applies to all Vetehine models and LoRAs, whether or not this notice is explicitly included with a specific model or release.