Dettagli
Raccomandazioni
1.7D
v1.7S
v1.6D
v1.6S
v1.5D
v1.5S
v1.41D
v1.41S
v1.4S
v1.4
v1.35
v1.35 GGUF Q4_K_M
v1.34
v1.34 GGUF Q4_K_M
v1.34_8Step
v1.33_8step
v1.32 8step
v1.32
v1.31 8step
v1.31
Moxie Fusion Flux

Moxie Fusion Flux

395
38
401
#futuristic
#นิยายวิทยาศาสตร์
#Cyberpunk
#scifi
#cybernetic
#Base Model
#ไซไฟ
#เรียลลิสติก
#Mechanical
#robotic
#vikingpunk
#biopunk
#photorealisic
#implants
#มายธ์พังค์
#flux1.ง
#augments
#Westernpunk
#นินจาปังค์
#เครื่องจักรกล_แขน
#อาวองการ์ดพังก์
#FLUX

v1.7

v1.6

This is the list of current trigger words for the embedded Lora (works great w/o any - of course); biopunk, youngMoxie, westernpunk, western_punk, vikingpunk, viking_punk, celticpunk, ninjapunk, ninja_punk, mythpunk, mechanical_arms, avant_garde_punk, avantGardePunk, augments, implants,

NOTE: Versions prior to v1.6 (IE 1.5S and below) included Moxie Punk Flux D in all versions and follow flux-dev licensing. However, v1.6S is includes Moxie Punk v3.0S which is 100% S - with no Dev inclusion.

v1.5

v1.41

  • 1.41D This is basically the same as v1.4 with a few exceptions (that I think give a big bump in quality). MoxieFlux1 DS v1.5D and tuned it with the newest Moxie Punk flux v1.1. However a smidgen of UltraReal Fine-Tune v2.0 was also merged.

  • 1.41S This is basically the same as v1.4S with a few exceptions (that I think give a big bump in quality). MoxieFlux1 DS v1.5S and tuned it with the newest Moxie Punk Flux v1.1.

1.4s

This is a simple merge of Shuttle Diffusion 3, Flux-dev2pro-fp8 special, and Moxie Punk. I use CFG 1.5, and 4 steps. A detailer or two is definitely helpful. Personally, I think the quality is pretty decent considering the fast generation times.

1.4

This is a simple merge of C4pacitor v.d_v2, and Moxie Punk.

1.35v1.51D

This is a fine tune of Moxie Fusion Flux v1.41D using Moxie Punk flux v2.

Added to this version are a new v1.4 base (working on getting it posted), and NinjaPunk and WesterPunk Lora's. It seems to have more variety in the 'female appearance' so far. Also, it is very viable without using any trigger words. So far, I am quite pleased with the quality of the images.

From here, I will probably start working on robotic, and overall 'cyberpunk' - which will be a new major release. While I have always done cyberpunk type stuff, this will be a new venture for me.

1.34

'Full' version is 'redone' with e5m2 fp16, and not hyper Lora. However, I left the stillsuit out. GGUF is based upon this version.

1.34 adds an unpublished version of the augments, and the new stillsuit has also been included. I am quite pleased with the results. As I improve the other punk styles, they will be merged in, and further released of Moxie Flux Fusion will be released.

NOTE: This checkpoint functions well with no use of trigger words.

1.33

NF4 version is available for download under Files for v1.33

I have been working on 1.33 for a while. I finally was able to get Lora's that seemed to work well and give me clean images that I was able to merge that gave me what I was looking for. However, this is DEFINITELY experimental still. It still spits out some blurry and deformed faces etc. periodically. I do like a lot of the images it is outputting, though.

Merged into this are Mechanical Arms v1, not so Post Apocalypse v1, and VikingPunk v2. However, it definitely does not require trigger words.

  • v1.31 and 1.32 are f16 e5m2.

    • 1.31 is the base for 1.32

    • 1.32 definitely has some quality issues

  • 1.33 is fp8 e4m3fn

Note: v1.31 is the base model before my styles get added for v1.32.

Welcome to my first trained model! I merged multiple models, and a few Loras, to create the base I wanted. Once I achieved the desired aesthetic, I merged in my personal Lora which has been honed over countless iterations in ComfyUI. The final result is a checkpoint that I'm thrilled to share with the Civitai community.

To further refine the model, I carefully merged it back with the base dev at e5m2 f16. This forms the foundation for the punk stylings. For an extra layer of visual flair, I wove in Detailed Perfection Style (triggered by "perfection style") and Sinfully Stylish (detailed lighting), elevating the lighting and detail to new heights.

As I continue to craft increasingly potent Loras, you can expect a steady stream of updates to my Cybernetic and Punk series. Each update will be thoughtfully integrated into this checkpoint. Currently, only Augments v2 is merged in, which can be triggered by specifying "augments" or "implants" in your prompts.

While I've made both the full checkpoint and an 8-step variant available, my older eyes struggle to discern a noticeable difference between the two. Feel free to experiment with both and decide which suits your needs best.

Happy diffusion, and stay punk!

Visualizza la traduzione

Valutazioni e recensioni

5.0 /5
0 valutazioni

Non ancora ricevute valutazioni o commenti sufficienti

no-data
Nessun dato disponibile
M
Conversazione con il modello
Annuncio
2024-09-21
Pubblicare modello
2025-05-05
Aggiorna le informazioni del modello
Dettagli modello
Tipo
Checkpoint
Data di pubblicazione
2025-04-30
Modello Base
Flux.1 D
Introduzione alla versione

v1.7

Ambito di Licenza
Model Source: civitai

1. I diritti dei modelli ripubblicati appartengono ai creatori originali.

2. I creatori originali che desiderano reclamare il proprio modello devono contattare il personale di SeaArt AI tramite canali ufficiali. Clicca per rivendicare

Ambito di Licenza di Creazione
Immagini Online
Effettua una Fusione
Consenti Download
Licenza Commerciale
Le immagini generate possono essere vendute o utilizzate per scopi commerciali
Consenti la rivendita del modello o la vendita dopo la fusione
QR Code
Scarica l'app SeaArt
Continua il tuo viaggio di creazione AI su mobile