---
license: other
license_name: flux-1-dev-non-commercial-license
license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
language:
- en
tags:
  - text-to-image
  - stable-diffusion
  - safetensors
  - stable-diffusion-xl
---
<style>
  .title{
    font-size: 2.5em;
    letter-spacing: 0.01em;
    padding: 0.5em 0;
  }
	.thumbwidth{
	max-width: 180px;
	}
	.font_red{
		color:red
	}
</style>

## Recent Updates
24/10/01 (Check Point) [myjc_flux_v2 Finetune Test](#myjc)
24/09/26 (LoRA) [myjc_flux_lora_v2-test](#myjc)

---

<a id="myjk"></a> 
<h1 class="title">
<span>myjk flux</span>
</h1>
 -trained 2852+1316 images.<br/>
 -The trigger doesn't seem valid...<br/>
 <br/>
 <br/>
 
[Download: myjk_flux_lora_v1](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjk_flux_lora_v1.safetensors?download=true) (LoRA)<br/>
[Download: myjk_flux-Q5_K_M.gguf](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjk_flux-Q5_K_M.gguf?download=true) (checkpoint)<br/>
[Download: version b](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjk_flux_b-Q5_K_M.gguf?download=true) (+109 images use adamwschedulefree optimizer)<br/>
<br/>
## Recommended:<br/>
The LoRA used for the test is [Flux Fusion DS v0 GGUF Q5_K_M](https://civitai.com/models/630820?modelVersionId=765575).
<br/>
VAE / Text Encoder: ae, clip_l, t5-v1_1-xxl-encoder-Q5_K_M<br/>
<table>
<tr>
<td>
<a href="https://img99.pixhost.to/images/705/514290586_20240920163149_myjk_flux-q5_k_m_1769369977.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/705/514290586_20240920163149_myjk_flux-q5_k_m_1769369977.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/705/514290592_20240920170058_myjk_flux-q5_k_m_872243841.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/705/514290592_20240920170058_myjk_flux-q5_k_m_872243841.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/705/514293472_20240920174336_myjk_flux-q5_k_m_2913518537.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/705/514293472_20240920174336_myjk_flux-q5_k_m_2913518537.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>

</table>
-refer to png info
<br />

## - sample prompt
[<img src=https://t99.pixhost.to/thumbs/705/514290595_20240920171937_myjk_flux-q5_k_m_3220485898.jpg />](https://img99.pixhost.to/images/705/514290595_20240920171937_myjk_flux-q5_k_m_3220485898.jpg)
```bash
japanese, 18yo, myjk, smile,
photograph of Two girls in idol costumes singing. The girl on the left has black ponytail hair and a guitar. The girl on the right has long black hair and a microphone. The stage at night is illuminated with lights and neon “myjk” signage.
Steps: 12, Sampler: Euler, Schedule type: Simple, CFG scale: 1, Seed: 3220485898, Size: 768x1024, Model hash: 33c0966fb8, Model: myjk_flux-Q5_K_M, Denoising strength: 0.3, Hires CFG Scale: 1, Hires upscale: 2, Hires upscaler: 4x-UltraSharp, Version: f2.0.1v1.10.1-previous-535-gb20cb4bf0, Diffusion in Low Bits: Automatic (fp16 LoRA), Module 1: ae, Module 2: t5-v1_1-xxl-encoder-Q5_K_M, Module 3: clip_l
```
<br />

## - trigger
```bash
myjk, japanese, european, 
and 16-18 yo, 
and native english(recomended) or danbooru tags
```
<br/>


<a id="myjc"></a> 
<h1 class="title">
<span>myjc flux</span>
</h1>
 -trained 1543+1309 images.<br/>
 -The trigger doesn't seem valid...<br/>
 <br/>
 <br/>
 v2<br />
 
 [Download: myjc_flux_lora_v2-test](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjc_flux_lora_v2-test.safetensors?download=true) (LoRA) #Flux-Dev2_Pro training Test
 <span class="font_red">If the image is blurred, increase Sampling steps</span><br>
 
 [Download: myjc_flux_v2_FTtest-Q5_K_M](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjc_flux_v2_FTtest-Q5_K_M.gguf?download=true) (Check Point) #Flux dev1 Finetune + LoRA Test <br/>
 <br />
 v1<br />
[Download: myjc_flux_lora_v1](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjc_flux_lora_v1.safetensors?download=true) (LoRA)<br/>
[Download: myjc_flux-Q5_K_M.gguf](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjc_flux-Q5_K_M.gguf?download=true) (checkpoint)<br/>
<br/>
## Recommended:<br/>
The LoRA used for the test is [Flux Fusion DS v0 GGUF Q4_0 (UNET)](https://civitai.com/models/630820?modelVersionId=736086) and [v0 GGUF Q5_K_M](https://civitai.com/models/630820?modelVersionId=765575).
<br/>
VAE / Text Encoder: ae, clip_l, t5-v1_1-xxl-encoder-Q5_K_M<br/>
<table>
<tr>
<td>
<a href="https://img99.pixhost.to/images/338/509944057_20240904212108_myjc_flux-q5_k_m_803013794.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/338/509944057_20240904212108_myjc_flux-q5_k_m_803013794.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/338/509944058_20240904214557_myjc_flux-q5_k_m_2287512062.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/338/509944058_20240904214557_myjc_flux-q5_k_m_2287512062.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/338/509944061_20240904220631_myjc_flux-q5_k_m_3636763026.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/338/509944061_20240904220631_myjc_flux-q5_k_m_3636763026.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>

</table>
-refer to png info
<br />

## - sample prompt
[<img src=https://t99.pixhost.to/thumbs/338/509944238_20240905080824_myjc_flux-q5_k_m_1298706659.jpg />](https://img99.pixhost.to/images/338/509944238_20240905080824_myjc_flux-q5_k_m_1298706659.jpg)
```bash
14yo, myjc, japanese, medium breasts,
This photograph captures a girl sitting on a grassy field at night. She has a light complexion and straight long black hair with bangs styled with a black bow. Her expression is cheerful with a slight smile. She is wearing a loose oversized shirt in a pastel gradient of pink yellow and blue which is slightly oversized giving it a cozy casual look. Her shirt is paired with white shorts and knee-high black socks with a small white bow on the top. The socks are adorned with a subtle pattern. She sits on a blanket with a white background featuring small amo,e characters. The grass is lush and green indicating a well-maintained lawn. The background is dark suggesting it is nighttime and the lighting is soft creating a warm and intimate atmosphere. The overall mood of the image is relaxed and playful with the subject's youthful and cheerful demeanor complementing the serene outdoor setting.
Steps: 12, Sampler: Euler, Schedule type: Simple, CFG scale: 1, Seed: 1298706659, Size: 768x1024, Model hash: c6b19f170d, Model: myjc_flux-Q5_K_M, Denoising strength: 0.3, Hires upscale: 2, Hires upscaler: 4x-UltraSharp, Version: f2.0.1v1.10.1-previous-501-g668e87f92, Diffusion in Low Bits: Automatic (fp16 LoRA), Module 1: ae, Module 2: clip_l, Module 3: t5-v1_1-xxl-encoder-Q5_K_M
```
<br />

## - trigger
```bash
myjc, japanese, european, 
and 13-15 yo, 
and native english(recomended) or danbooru tags
```
<br/>

---

<a id="test03"></a> 
<h1 class="title">
<span>lora_zipang_flux_test</span>
</h1>
-Training was based on a merged model of dev1 and lora test**.<br/>
<br/>

### -Trigger
```bash
japanese, european
```
<br/>

* [test04](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/lora_zipang_flux_test04.safetensors?download=true) +350 images
```bash
myjc, 13yo
```

* [test03](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/lora_zipang_flux_test03.safetensors?download=true) +920 images
```bash
myjsh, 12yo
```

 <br/>
<a id="test02"></a> 
<h1 class="title">
<span>myjsm_flux_test02</span>
</h1>
  -It is a test lora of poor quality with only a few images learned.<br/>
 -trained 273 images.<br/>
  <br/>
Found a slightly better training setting.
But still hard to find things that don't show up in flux.
 <br/>
 <br/>
 
[Download:test02](https://huggingface.co/deadman44/Flux_Photoreal_LoRA/resolve/main/myjsm_flux_test02.safetensors?download=true) <br/>
<br/>
The model used for the test is [Flux Fusion DS v0 GGUF Q4_0 (UNET)](https://civitai.com/models/630820?modelVersionId=736086) and [v0 GGUF Q5_K_M](https://civitai.com/models/630820?modelVersionId=765575).
<table>
  <tr>
    <td colspan="3">
      <div>
GGUF Q4_0 + t5xxl_fp8_e4m3fn : 4step
      </div>
	  </td>
	</tr>
<tr>
<td>
<a href="https://img99.pixhost.to/images/126/507626249_20240827094724_fusionds_v0_q4_456078958.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626249_20240827094724_fusionds_v0_q4_456078958.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626251_20240827103511_fusionds_v0_q4_482040669.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626251_20240827103511_fusionds_v0_q4_482040669.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626253_20240827112528_fusionds_v0_q4_1816421730.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626253_20240827112528_fusionds_v0_q4_1816421730.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>
  <tr>
    <td colspan="3">
      <div>
GGUF Q5_K_M. + t5-v1_1-xxl-encoder-Q5_K_M : 12step
      </div>
	  </td>
	</tr>
<tr>
<td>
<a href="https://img99.pixhost.to/images/126/507626250_20240827102458_fluxfusionds_v0_q5_k_m_2418428235.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626250_20240827102458_fluxfusionds_v0_q5_k_m_2418428235.jpg" alt="sample1"  class="thumbwidth" >
      </div>
    </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626252_20240827110802_fluxfusionds_v0_q5_k_m_3216545735.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626252_20240827110802_fluxfusionds_v0_q5_k_m_3216545735.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
<td>
<a href="https://img99.pixhost.to/images/126/507626256_20240827121409_fluxfusionds_v0_q5_k_m_2982180625.jpg" target=”_blank”>
      <div>
<img src="https://t99.pixhost.to/thumbs/126/507626256_20240827121409_fluxfusionds_v0_q5_k_m_2982180625.jpg" alt="sample1"  class="thumbwidth" >
      </div>
	  </td>
  </tr>
</table>
-refer to png info
<br />

## - sample prompt
[<img src=https://t99.pixhost.to/thumbs/126/507626257_20240827124249_fusionds_v0_q4_642879771.jpg />](https://img99.pixhost.to/images/126/507626257_20240827124249_fusionds_v0_q4_642879771.jpg)
```bash
9yo, myjsm, japanese,
photograph of a girl sitting on a brick pavement with a pink umbrella in front of her. She is wearing a white camisole and a blue skirt with a anime print. She has shoulder-length dark hair and is smiling at the camera.
bangs, black eyes, skirt, rain
<lora:myjsm_flux_test02:1>
Steps: 4, Sampler: Euler, Schedule type: Simple, CFG scale: 1, Seed: 642879771, Size: 792x1056, Model hash: 5e21feb505, Model: FusionDS_v0_Q4, Lora hashes: "myjsm_flux_test02: 3fdff20b7d65", Version: f2.0.1v1.10.1-previous-419-gf82029c5c, Module 1: ae, Module 2: clip_l, Module 3: t5xxl_fp8_e4m3fn
```
<br />

## - trigger
```bash
myjsm, japanese, 9yo, 
and native english
```
<br />

## -Train Settings
```bash
base model: flux1-dev.safetensors
vae/text encoder: clip_l.safetensors, t5xxl_fp8_e4m3fn.safetensors, ae.safetensors
tag: caption (native eng) + tags (danbooru)

--network_module "networks.lora_flux"
--gradient_checkpointing
--cache_latents
--cache_latents_to_disk
--cache_text_encoder_outputs
--cache_text_encoder_outputs_to_disk
--enable_bucket
--bucket_no_upscale
--optimizer_type "AdamW8bit"
--optimizer_args "weight_decay=0.01" "betas=0.9,0.999"
--learning_rate=0.0002
--network_dim=32
--network_alpha=4
--network_train_unet_only
--mixed_precision "bf16"
--save_precision "bf16"
--full_bf16
--loss_type "l2"
--huber_schedule "snr"
--model_prediction_type "raw"
--discrete_flow_shift 3
--timestep_sampling "sigma"
--max_grad_norm=1
--max_timestep=1000
--min_snr_gamma=5
--min_timestep=100
--noise_offset=0.0375
--adaptive_noise_scale=0.00375
--apply_t5_attn_mask
--split_mode
--network_args "loraplus_unet_lr_ratio=16" "train_blocks=single"
```
<br />