Upload 12 files

#/1980s_Fantasy_Movies_HunyuanVideo_Lora/hunyuan_80s_fantasyv1_5_1000_diffusers.md

@@ -0,0 +1,45 @@
+# 1980s Fantasy Movies HunyuanVideo Lora (Version: v1.5 Diffusers)
+
+**Creator**: [neph1](https://civitai.com/user/neph1)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-05-21
+**Trigger Words**: `80s_fantasy`
+
+**Civitai Model ID**: 1386261
+**Civitai Version ID**: 1815189
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 27
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+The lora I was afraid to make. Not because I didn't want it, but I was afraid of failing. So much of my childhood is contained in this lora. It's trained on images < 400p and will lose strength the higher you push the resolution. I've mostly tried conditioning between 8 and 10. You don't really need the trigger word, imo, but it could be used to channel some more style into the generation Does not do so well with combat scenes. Body horror is likely to ensue. For an update I will add some appropriate clips and see if that helps. For prompt hints, check the example generations I've posted. Here's also one from the training data, to give you a sense of the format: "A group of robed figures, likely a council of wizards or sorcerers, stand around a glowing, pink crystal at the center of a circular stone table. The crystal emits a mystical aura, and the figures appear to be discussing or casting a spell. The background features a dark, cavernous interior with stone walls and alcoves containing various objects, including a large, skeletal creature. The setting suggests a magical or ancient ruin."
+
+## Version Notes (v1.5 Diffusers)
+
+To complete the set. Don't know why I didn't upload this, before.
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1386261?modelVersionId=1815189)**
+*   [View Full Model Page →](https://civitai.com/models/1386261)
+*   [View Creator Profile →](https://civitai.com/user/neph1)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `hunyuan_80s_fantasyv1_5_1000_diffusers.safetensors`
+*   **Filename (Civitai)**: `hunyuan_80s_fantasyv1_5_1000_diffusers.safetensors`
+*   **Size**: 312.05 MB
+*   **Hash (AutoV2)**: `D47F8A111A`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/1980s_Fantasy_Movies_HunyuanVideo_Lora/hunyuan_80s_fantasyv1_5_1000_diffusers.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d47f8a111a2c99beab6d1baa977f28616d7e0e3ddbc656a27234c20671b1634e
+size 327212096

#/1980s_Horror_movie_Hunyuan_Video_lora/hunyuan_80s_horror_1000.md

@@ -0,0 +1,45 @@
+# 1980s Horror movie Hunyuan Video lora (Version: v1.0 - Diffusers Lora)
+
+**Creator**: [neph1](https://civitai.com/user/neph1)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-05-18
+**Trigger Words**: `80s_horror`
+
+**Civitai Model ID**: 1592586
+**Civitai Version ID**: 1802260
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 117
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+Bring out your werewolf costumes and put on the zombie makeup. Set "Thriller" on repeat. If you thought body horror was bad with video models before, it'll be much worse with this lora! Captioning this was tricky, as the automatic captioning had troubles with the 'variety' of the images here. So, a lot of hands-on required to clean it up. It's a sub-par dataset. I noticed afterwards that I had left some vignettes in. It's also fairly small, and I've left out many popular movies. Room for improvement. Example captions from the dataset: "A man with a gaunt, pale face, sunken eyes, and thin lips. He has short, disheveled hair and is wearing a dark, stained jacket over a red shirt. He is pointing at the viewer with his right hand, which is also stained. The background is a concrete wall with visible cracks and splatters of paint in various shades of green, brown, and gray. The camera angle is straight-on, capturing the man from the front." "A man in a tattered, white robe stands in a lush, tropical jungle. His face is obscured by a hood. In the foreground is a tombstone with a cross on it. The background is dense with greenery and palm trees, creating a dark, mysterious atmosphere. The camera angle is low, looking up at the man, emphasizing his imposing presence. The lighting is dim, casting shadows and adding to the eerie, ominous mood."
+
+## Version Notes (v1.0 - Diffusers Lora)
+
+Guidance: 10
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1592586?modelVersionId=1802260)**
+*   [View Full Model Page →](https://civitai.com/models/1592586)
+*   [View Creator Profile →](https://civitai.com/user/neph1)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `hunyuan_80s_horror_1000.safetensors`
+*   **Filename (Civitai)**: `hunyuan_80s_horror_1000.safetensors`
+*   **Size**: 312.05 MB
+*   **Hash (AutoV2)**: `B9D23DDCC2`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/1980s_Horror_movie_Hunyuan_Video_lora/hunyuan_80s_horror_1000.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9d23ddcc2351015722ebe88b5799967b46ff8acf3ab61c80f534f62824046b7
+size 327212096

#/1980s_Horror_movie_Hunyuan_Video_lora/hunyuan_80s_horror_1000_converted.md

@@ -0,0 +1,45 @@
+# 1980s Horror movie Hunyuan Video lora (Version: v1.0 - Comfy compatible)
+
+**Creator**: [neph1](https://civitai.com/user/neph1)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-05-18
+**Trigger Words**: `80s_horror`
+
+**Civitai Model ID**: 1592586
+**Civitai Version ID**: 1804220
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 63
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+Bring out your werewolf costumes and put on the zombie makeup. Set "Thriller" on repeat. If you thought body horror was bad with video models before, it'll be much worse with this lora! Captioning this was tricky, as the automatic captioning had troubles with the 'variety' of the images here. So, a lot of hands-on required to clean it up. It's a sub-par dataset. I noticed afterwards that I had left some vignettes in. It's also fairly small, and I've left out many popular movies. Room for improvement. Example captions from the dataset: "A man with a gaunt, pale face, sunken eyes, and thin lips. He has short, disheveled hair and is wearing a dark, stained jacket over a red shirt. He is pointing at the viewer with his right hand, which is also stained. The background is a concrete wall with visible cracks and splatters of paint in various shades of green, brown, and gray. The camera angle is straight-on, capturing the man from the front." "A man in a tattered, white robe stands in a lush, tropical jungle. His face is obscured by a hood. In the foreground is a tombstone with a cross on it. The background is dense with greenery and palm trees, creating a dark, mysterious atmosphere. The camera angle is low, looking up at the man, emphasizing his imposing presence. The lighting is dim, casting shadows and adding to the eerie, ominous mood."
+
+## Version Notes (v1.0 - Comfy compatible)
+
+The same as the diffusers version, but compatible with ComfyUI without the need for custom code or nodes.
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1592586?modelVersionId=1804220)**
+*   [View Full Model Page →](https://civitai.com/models/1592586)
+*   [View Creator Profile →](https://civitai.com/user/neph1)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `hunyuan_80s_horror_1000_converted.safetensors`
+*   **Filename (Civitai)**: `hunyuan_80s_horror_1000_converted.safetensors`
+*   **Size**: 1170.02 MB
+*   **Hash (AutoV2)**: `E41BE870AC`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/1980s_Horror_movie_Hunyuan_Video_lora/hunyuan_80s_horror_1000_converted.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e41be870ac26f83c8a0eead9746c6247e9db31f9feb8064fa71d5edac08726f5
+size 1226854896

#/3D-Realistic_Doggystyle_Anal_(Hunyuan)/TifaAnal-000008.md

@@ -0,0 +1,41 @@
+# 3D/Realistic Doggystyle Anal (Hunyuan) (Version: V1)
+
+**Creator**: [Felldude](https://civitai.com/user/Felldude)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-05-13
+**Trigger Words**: `Doggy Style, Doggy Style Anal Sex, Fucked From behind, 3d, realistic `
+
+**Civitai Model ID**: 1574426
+**Civitai Version ID**: 1781658
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 299
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+3D/Realistic Doggystyle Anal Trained with full 3D images not just frames
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1574426?modelVersionId=1781658)**
+*   [View Full Model Page →](https://civitai.com/models/1574426)
+*   [View Creator Profile →](https://civitai.com/user/Felldude)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `TifaAnal-000008.safetensors`
+*   **Filename (Civitai)**: `TifaAnal-000008.safetensors`
+*   **Size**: 225.09 MB
+*   **Hash (AutoV2)**: `05ED806153`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/3D-Realistic_Doggystyle_Anal_(Hunyuan)/TifaAnal-000008.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05ed80615393a9cfd5be9d968d3621bdcff0886571ce4a87bb6cc3a25b371962
+size 236019040

#/80's_porn_centerfold/pornstar_v2_epoch8.md

@@ -0,0 +1,45 @@
+# 80's porn centerfold (Version: v1.0)
+
+**Creator**: [KinkMaster](https://civitai.com/user/KinkMaster)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-03-15
+**Trigger Words**: `a 1980's porn centerfold`
+
+**Civitai Model ID**: 1361239
+**Civitai Version ID**: 1537790
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 377
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+Just a fun little lora experiment that I spent a bit too much time on. I trained this three times, which is crazy considering it was meant as a one and done dump. But I learned a good amount of information along the way. Trained on 54 still images of porn centerfold models from the 1980s. Intent was to capture: 1. the general look of the women in that time period, and 2. lighting and feel of the video/photography used for porn magazines then. For a quick side lora (or what intended to be quick) I think it does well enough. Because of the still image use only, I've found that this lora does not play well with higher frame counts. 89 frames is the sweet spot. 121 will still have movement, but it will be diminished. 145 frames will almost always result in still images only. Don't waste your time. Captioning was extremely simple: A 1980's porn centerfold woman You can basically prompt the above with some sort of generic pose and location and the lora will do a decent job of producing a woman with big hair (both on her head and down below). Give it a try and have fun. And for anyone who doesn't care about training data, then stop here. V1 - Trained it with an aim of 1600 steps for 8 total epochs at learning rate (LR) of 8e5. You can see by the loss graph that it hit steep decline by epoch 2 only to rise up and back down more slowly for epoch 4 before taking off and overfitting in epochs 5-8. V2 - I immediately threw prior results out and trained for 10 Epochs at LR 7e5 instead to try to avoid overfitting the so quickly. The lower LR seemed to work better, though the ramp down on epoch 6 was rather steep. It too was followed by a spike and then another dip before climbing steadily toward being overfit. When testing between Epoch 6 and 8, Epoch 8 actually showed more adherence to prompts and general cohesion toward the premise trained without noticeable harm to motion. It was during these test inferences that I noticed the diminished motion at the higher frame counts. V3 - As an experiment, I tried modifying captions to say "An image of..." at the beginning and then trained it on the exact same settings as before. I read on discord that this can sometimes help the model understand that the concepts are being trained off static images and thus avoid some of the fighting that an inherently overtrained lora can do against motion. The downward fall was steady and quick for this followed by a slow and steady overtraining that then resulted in another long curve back down. After testing, the Epoch 2 and 10 loras that were trained were not coherent enough and resulted in far more artifacts than Epoch 8 under V2. As a result, I have chosen Epoch 8 of V2 as the winner. This does show the importance of captioning in lora training though, at least from my perspective. And there is something to be said about the use of "an image of" and "a video of" used in dual image/video datasets. Stand-alone for a single medium, it is probably not very needed. Another take away for this is to not throw away epochs after a slight climb in the loss. Sometimes the epoch after can still be useful. Ultimate take away is, if you see mid-training climb in loss, you may want to try lowering the LR to offset and help stabilize a bit.
+
+## Version Notes (v1.0)
+
+54 images trained at 5e8
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1361239?modelVersionId=1537790)**
+*   [View Full Model Page →](https://civitai.com/models/1361239)
+*   [View Creator Profile →](https://civitai.com/user/KinkMaster)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `pornstar_v2_epoch8.safetensors`
+*   **Filename (Civitai)**: `pornstar_v2_epoch8.safetensors`
+*   **Size**: 172.55 MB
+*   **Hash (AutoV2)**: `707B0A72E7`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/80's_porn_centerfold/pornstar_v2_epoch8.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:707b0a72e7ae63409019754d89b0ca8f9d1d2545924a056fc4ccefa3144449fc
+size 180930896

#/90s_MTV_The_Grind_Hunyuan_Video_Lora/hunyuan_mtv_grind_500.md

@@ -0,0 +1,41 @@
+# 90s MTV The Grind Hunyuan Video Lora (Version: v1.0 - Diffusers Lora)
+
+**Creator**: [neph1](https://civitai.com/user/neph1)
+**Type**: LORA
+**Base Model**: Hunyuan Video
+**Version Uploaded**: 2025-05-13
+**Trigger Words**: `mtv_grind, dancing`
+
+**Civitai Model ID**: 1577139
+**Civitai Version ID**: 1784685
+
+**Stats (at time of fetch for this version)**:
+*   Downloads: 54
+*   Rating: 0 (0 ratings)
+*   Favorites: N/A
+
+---
+
+## 📄 Description (Parent Model)
+
+This is not your average sexy dance lora. It didn't turn out quite the way I wanted due to the extremely low quality source material I could find. And yet, it does what it was trained to do. If I can find better quality video, and people enjoy it, I'll update it. I recommend running with teacache off, as it has a tendency to mess up the limbs of the dancers. Sample prompt: "a group of young people in, dancing energetically. the setting is outside in a pool area with a blue sky above. in the background we see palm trees. the camera is slightly from above, zooming in on one man as he dances." In general, all my lora captions (with some exceptions or variations) follow this format: central subject. action of subject. setting / background. camera.
+
+---
+
+## Civitai Links
+
+*   **[🔗 View This Version on Civitai →](https://civitai.com/models/1577139?modelVersionId=1784685)**
+*   [View Full Model Page →](https://civitai.com/models/1577139)
+*   [View Creator Profile →](https://civitai.com/user/neph1)
+
+---
+
+## File Information
+
+*   **Filename (Saved Locally)**: `hunyuan_mtv_grind_500.safetensors`
+*   **Filename (Civitai)**: `hunyuan_mtv_grind_500.safetensors`
+*   **Size**: 432.07 MB
+*   **Hash (AutoV2)**: `DD77E39BBC`
+
+---
+*Downloaded locally using CivitaiLocalDownloader/1.4-SpecificVersions-NoCreatorFolder.*

#/90s_MTV_The_Grind_Hunyuan_Video_Lora/hunyuan_mtv_grind_500.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd77e39bbcf82fafdd197eb3f8e052d67980b431cb59ef1a23548485c995ae79
+size 453063056