Did you lose your beloved pet recently? Fear not, now you too can immortalize it forever by using this simple taxidermy model.Use the trigger word: "badtxdrmy, bad taxidermy" along with the LoRA to activate it. Since it's Flux, you will want to describe the animal and scene carefully.The model was inspired by <a rel="ugc" href="https://civarchive.com/models/855329/bad-taxidermy-for-flux?modelVersionId=983653">Bad Taxidermy for Flux</a> by <a rel="ugc" href="https://civarchive.com/user/Erentil">Erentil</a>.The dataset was made entirely by <a rel="ugc" href="https://civarchive.com/user/526christian">526christian</a>. Thank you for sharing &lt;3<h3 id="example-prompt-fgr4biyzj">Example Prompt</h3><pre><code>A photo of a badtxdrmy, bad taxidermy, A poorly taxidermied opossum, its beady eyes crossed and its fur matted, hangs from a rusty metal hook in a dimly lit, dusty antique shop. The opossum's mouth is stretched into a comically exaggerated snarl, revealing crooked, yellowed teeth. Its prehensile tail has been awkwardly attached to its backside, sticking straight up in the air like a furry satellite dish. 

The opossum appears to be "dressed up" for a tea party, wearing a tiny, ill-fitting top hat and a faded, floral-patterned waistcoat that's several sizes too small. A delicate china teacup is clutched in its paw, which looks like it's about to shatter at any moment. The overall effect is one of charmingly inept taxidermy and a delightfully absurd scenario.&lt;lora:BadTaxidermiFlux:1&gt;</code></pre><h3 id="prompting-with-llm-oolnhot4u">Prompting with LLM</h3>I'm using an LLM in Comfy to craft the prompts used for this model. I'm using my <a target="_blank" rel="ugc" href="https://github.com/MNeMoNiCuZ/ComfyUI-mnemic-nodes?tab=readme-ov-file#-groq-llm-api-node">Groq node pack</a>.Here's the prompt I'm using with the LLM:<pre><code>[INSTRUCTION]Create a unique image generation prompt for a "bad taxidermi" image. The ones where the animal is slightly disfigured and ugly. Also describe the scene where the animal is. Do not use words or writing as part of the prompt.

Do not acknowledge the user. Only return a visual description based on the user's input.

User input:</code></pre><h3 id="generation-settings-qukgi1nwf">Generation Settings</h3>It is flux, everything seems to work fine for me. I use 20 steps, dpmpp_2m sampler with beta scheduler.<h2 id="tips-and-tricks-21afevkpg">Tips &amp; Tricks</h2>Really have at describing the scene well!

Bad Taxidermy Animals [Flux]

BadTaxidermiFlux.safetensors

v1.0

Did you lose your beloved pet recently? Fear not, now you too can immortalize it forever by using this simple taxidermy model.Use the trigger word: "badtxdrmy, bad taxidermy" along with the LoRA to activate it. Since it's Flux, you will want to describe the animal and scene carefully.The model was inspired by <a target="_blank" rel="ugc" href="https://civitai.com/models/855329/bad-taxidermy-for-flux?modelVersionId=983653">Bad Taxidermy for Flux</a> by <a target="_blank" rel="ugc" href="https://civitai.com/user/Erentil">Erentil</a>.The dataset was made entirely by <a target="_blank" rel="ugc" href="https://civitai.com/user/526christian">526christian</a>. Thank you for sharing &lt;3Example PromptA photo of a badtxdrmy, bad taxidermy, A poorly taxidermied opossum, its beady eyes crossed and its fur matted, hangs from a rusty metal hook in a dimly lit, dusty antique shop. The opossum's mouth is stretched into a comically exaggerated snarl, revealing crooked, yellowed teeth. Its prehensile tail has been awkwardly attached to its backside, sticking straight up in the air like a furry satellite dish. The opossum appears to be "dressed up" for a tea party, wearing a tiny, ill-fitting top hat and a faded, floral-patterned waistcoat that's several sizes too small. A delicate china teacup is clutched in its paw, which looks like it's about to shatter at any moment. The overall effect is one of charmingly inept taxidermy and a delightfully absurd scenario.&lt;lora:BadTaxidermiFlux:1&gt;Prompting with LLMI'm using an LLM in Comfy to craft the prompts used for this model. I'm using my <a target="_blank" rel="ugc" href="https://github.com/MNeMoNiCuZ/ComfyUI-mnemic-nodes?tab=readme-ov-file#-groq-llm-api-node">Groq node pack</a>.Here's the prompt I'm using with the LLM:[INSTRUCTION]Create a unique image generation prompt for a "bad taxidermi" image. The ones where the animal is slightly disfigured and ugly. Also describe the scene where the animal is. Do not use words or writing as part of the prompt. Do not acknowledge the user. Only return a visual description based on the user's input. User input:Generation SettingsIt is flux, everything seems to work fine for me. I use 20 steps, dpmpp_2m sampler with beta scheduler.Tips &amp; TricksReally have at describing the scene well!

Files with SHA256: eec2a8c8eb639f2918b08368725c9b60fc76ab8831107989741b688947f1562b