<h1 id="aesthetic-quality-modifiers-masterpiece">Aesthetic Quality Modifiers - Masterpiece</h1>Training data is a subset of all my manually rated datasets with the quality/aesthetic modifiers, including only the <code>masterpiece</code> tagged images.ℹ️ LoRA work best when applied to the base models on which they are trained. Please read the About This Version on the appropriate base models, trigger usage, and workflow/training information.Version 5.0 [anima-preview-3] (Latest)(Temporarily including here as the "About This Version" section is having issues)Trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/2458426/anima-official?modelVersionId=2836417">Anima Preview-3-base</a>Assume that any lora trained on the preview version won't work well on the final version.Recommended prompt structure:Positive prompt (quality tags at the start of prompt):<pre><code>masterpiece, best quality, very aesthetic, {{tags}}, {{natural language}}</code></pre>Updated dataset of 386 images, all masterpiece tagged images trained in <a target="_blank" rel="ugc" href="https://civarchive.com/models/2495369?modelVersionId=2898376">Kirazuri (Anima) model version 2</a> dataset.Trained at 1024 x 1024, 1280 x 1280, and 1536 x 1024 resolutions.Previews are mostly generated at 1536 x 1024 or 1024 x 1536 .Training config:diffusion-pipe commit b0aa4f1e03169f3280c8518d37570a448420f8be<pre><code># dataset-anima.toml
resolutions = [1024, 1280, 1536]

enable_ar_bucket = true
min_ar = 0.5
max_ar = 2.0
num_ar_buckets = 9

# Totals
# 386 images
# 15504 samples/epoch

# 153 images
# 48 samples/image - 7344 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1536x1536'
repeats = 16
resolutions = [1024, 1280, 1536]

# 44 images
# 48 samples/image - 2112 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1280x1280'
repeats = 24
resolutions = [1024, 1280]

# 189 images
# 32 samples/image - 6048 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1024x1024'
repeats = 32
resolutions = [1024]

# anima-lora.toml 
output_dir = '/mnt/d/anima/training_output/masterpieces-v5'

dataset = 'dataset-anima.toml'

# training settings
epochs = 5
# Per-resolution batch sizes
micro_batch_size_per_gpu = [[1024, 32], [1280, 24], [1536, 16]]
pipeline_stages = 1
gradient_accumulation_steps = 1
gradient_clipping = 1
warmup_steps = 100
lr_scheduler = 'cosine'

# misc settings
save_every_n_epochs = 1
activation_checkpointing = true

partition_method = 'parameters'

save_dtype = 'bfloat16'
caching_batch_size = 1
map_num_proc = 8
steps_per_print = 1
compile = true

[model]
type = 'anima'
transformer_path = '/mnt/c/workspace/models/diffusion_models/anima-preview3-base.safetensors'
vae_path = '/mnt/c/workspace/models/vae/qwen_image_vae.safetensors'
llm_path = '/mnt/c/workspace/models/text_encoders/qwen_3_06b_base.safetensors'
dtype = 'bfloat16'
llm_adapter_lr = 1e-6
flux_shift = true
multiscale_loss_weight = 0.5
sigmoid_scale = 1.3

[adapter]
type = 'lora'
rank = 32
dtype = 'bfloat16'

[optimizer]
type = 'adamw_optimi'
lr = 4e-5
betas = [0.9, 0.99]
weight_decay = 0.01
eps = 1e-8</code></pre>

Aesthetic Quality Modifiers - Masterpiece

Trained on <a target="_blank" rel="ugc" href="https://civitai.com/models/833294/noobai-xl-nai-xl?modelVersionId=1190596">NoobAI-XL (NAI-XL) V-Pred 1.0-Version</a>Recommended prompt structure:Positive prompt (quality tags at the end of prompt):<pre><code>{{tags}}
masterpiece, best quality, very aesthetic</code></pre>Included many newly rated images and small training/tagging updates:<ul><li>change<code>min_snr_gamma</code> from 5 to 0<ul><li>should produce closer lighting/blacks to base model</li></ul></li><li>change <code>very awa</code> tag back to <code>very aesthetic</code><ul><li>reverted this change from previous version v1.0</li></ul></li><li>tag <code>nsfw</code> <code>explicit</code> with imgutils anime_rating</li></ul>Previews are generated in Forge with DynamicThresholding (CFG-Fix) Integrated enabled, settings:<pre><code>dynthres_enabled: True, dynthres_mimic_scale: 7, dynthres_threshold_percentile: 1, dynthres_mimic_mode: Half Cosine Down, dynthres_mimic_scale_min: 1, dynthres_cfg_mode: Half Cosine Down, dynthres_cfg_scale_min: 3, dynthres_sched_val: 1, dynthres_separate_feature_channels: enable, dynthres_scaling_startpoint: ZERO, dynthres_variability_measure: STD, dynthres_interpolate_phi: 1</code></pre>A ComfyUI workflow reference with DynamicThresholding, Upscaling, and FaceDetailer can be found here: <a target="_blank" rel="ugc" href="https://civitai.com/posts/11457095">https://civitai.com/posts/11457095</a>

noobai_vpred_1_masterpieces_v23.safetensors

v2.3 [noobai-v-pred-1]

v5.0 [anima-preview-3]

v4.1 [anima-preview-2]

v4.0 [anima-preview]

v3.2 [z-image-turbo]

v3.1 [klein-4b]

v3.0 [illustrious]

v3.0 [qwen]

v2.0-alpha [wan-t2v-1.3b]

v2.0-alpha [wan-t2v-14b]

v1.0 [noobai-v-pred-1]

v1.0 [noobai-e-pred-1]

v1.0 [Illustrious]

v1.0 [Pony]

Aesthetic Quality Modifiers - MasterpieceTraining data is a subset of all my manually rated datasets with the quality/aesthetic modifiers, including only the masterpiece tagged images.Previews are generated in Forge with upscaling and adetailer.For Noobai V-Pred, a ComfyUI workflow reference with DynamicThresholding, Upscaling, and FaceDetailer can be found here: <a target="_blank" rel="ugc noopener noreferrer nofollow" href="https://civitai.com/posts/11457095">https://civitai.com/posts/11457095</a>Recommended prompt structure:Positive prompt (quality tags at the end of prompt):{{tags}} masterpiece, best quality, very aesthetic

&lt;p&gt;Trained on &lt;a target=&#34;_blank&#34; rel=&#34;ugc&#34; href=&#34;https://civitai.com/models/833294/noobai-xl-nai-xl?modelVersionId=1190596&#34;&gt;&lt;strong&gt;NoobAI-XL (NAI-XL) V-Pred 1.0-Version&lt;/strong&gt;&lt;/a&gt;&lt;/p&gt;&lt;p&gt;Included many newly rated images and small training/tagging updates:&lt;/p&gt;&lt;ul&gt;&lt;li&gt;&lt;p&gt;change&lt;code&gt;min_snr_gamma&lt;/code&gt; from 5 to 0&lt;/p&gt;&lt;ul&gt;&lt;li&gt;&lt;p&gt;should produce closer lighting/blacks to base model&lt;/p&gt;&lt;/li&gt;&lt;/ul&gt;&lt;/li&gt;&lt;li&gt;&lt;p&gt;change &lt;code&gt;very awa&lt;/code&gt; tag back to &lt;code&gt;very aesthetic&lt;/code&gt;&lt;/p&gt;&lt;ul&gt;&lt;li&gt;&lt;p&gt;reverted this change from previous version v1.0&lt;/p&gt;&lt;/li&gt;&lt;/ul&gt;&lt;/li&gt;&lt;li&gt;&lt;p&gt;tag &lt;code&gt;nsfw&lt;/code&gt; &lt;code&gt;explicit&lt;/code&gt; with imgutils anime_rating&lt;/p&gt;&lt;/li&gt;&lt;/ul&gt;&lt;p&gt;Previews are generated in Forge with DynamicThresholding (CFG-Fix) Integrated enabled, settings:&lt;/p&gt;&lt;pre&gt;&lt;code&gt;dynthres_enabled: True, dynthres_mimic_scale: 7, dynthres_threshold_percentile: 1, dynthres_mimic_mode: Half Cosine Down, dynthres_mimic_scale_min: 1, dynthres_cfg_mode: Half Cosine Down, dynthres_cfg_scale_min: 3, dynthres_sched_val: 1, dynthres_separate_feature_channels: enable, dynthres_scaling_startpoint: ZERO, dynthres_variability_measure: STD, dynthres_interpolate_phi: 1&lt;/code&gt;&lt;/pre&gt;

Aesthetic Quality Modifiers - MasterpieceTraining data is a subset of all my manally rated datasets with the quality/aesthetic modifiers, including only the masterpiece tagged images.Recommended prompt structure:Positive prompt (quality tags at the end of prompt):{{tags}} masterpiece, best quality, very aesthetic

Aesthetic Quality Modifiers Masterpiece - motimalu

v2.3[noobai-v-pre-1]

v1.0 [nooi-e-pred-1]

<h1 id="aesthetic-quality-modifiers-masterpiece">Aesthetic Quality Modifiers - Masterpiece</h1>Training data is a subset of all my manually rated datasets with the quality/aesthetic modifiers, including only the <code>masterpiece</code> tagged images.ℹ️ LoRA work best when applied to the base models on which they are trained. Please read the About This Version on the appropriate base models, trigger usage, and workflow/training information.Version 5.0 [anima-preview-3] (Latest)(Temporarily including here as the "About This Version" section is having issues)Trained on <a target="_blank" rel="ugc" href="https://civitai.red/models/2458426/anima-official?modelVersionId=2836417">Anima Preview-3-base</a>Assume that any lora trained on the preview version won't work well on the final version.Recommended prompt structure:Positive prompt (quality tags at the start of prompt):<pre><code>masterpiece, best quality, very aesthetic, {{tags}}, {{natural language}}</code></pre>Updated dataset of 386 images, all masterpiece tagged images trained in <a target="_blank" rel="ugc" href="https://civitai.red/models/2495369?modelVersionId=2898376">Kirazuri (Anima) model version 2</a> dataset.Trained at 1024 x 1024, 1280 x 1280, and 1536 x 1024 resolutions.Previews are mostly generated at 1536 x 1024 or 1024 x 1536 .Training config:diffusion-pipe commit b0aa4f1e03169f3280c8518d37570a448420f8be<pre><code># dataset-anima.toml
resolutions = [1024, 1280, 1536]

enable_ar_bucket = true
min_ar = 0.5
max_ar = 2.0
num_ar_buckets = 9

# Totals
# 386 images
# 15504 samples/epoch

# 153 images
# 48 samples/image - 7344 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1536x1536'
repeats = 16
resolutions = [1024, 1280, 1536]

# 44 images
# 48 samples/image - 2112 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1280x1280'
repeats = 24
resolutions = [1024, 1280]

# 189 images
# 32 samples/image - 6048 samples/epoch
[[directory]]
path = '/mnt/d/training_data/0_masterpieces_kirazuri/1024x1024'
repeats = 32
resolutions = [1024]

# anima-lora.toml 
output_dir = '/mnt/d/anima/training_output/masterpieces-v5'

dataset = 'dataset-anima.toml'

# training settings
epochs = 5
# Per-resolution batch sizes
micro_batch_size_per_gpu = [[1024, 32], [1280, 24], [1536, 16]]
pipeline_stages = 1
gradient_accumulation_steps = 1
gradient_clipping = 1
warmup_steps = 100
lr_scheduler = 'cosine'

# misc settings
save_every_n_epochs = 1
activation_checkpointing = true

partition_method = 'parameters'

save_dtype = 'bfloat16'
caching_batch_size = 1
map_num_proc = 8
steps_per_print = 1
compile = true

[model]
type = 'anima'
transformer_path = '/mnt/c/workspace/models/diffusion_models/anima-preview3-base.safetensors'
vae_path = '/mnt/c/workspace/models/vae/qwen_image_vae.safetensors'
llm_path = '/mnt/c/workspace/models/text_encoders/qwen_3_06b_base.safetensors'
dtype = 'bfloat16'
llm_adapter_lr = 1e-6
flux_shift = true
multiscale_loss_weight = 0.5
sigmoid_scale = 1.3

[adapter]
type = 'lora'
rank = 32
dtype = 'bfloat16'

[optimizer]
type = 'adamw_optimi'
lr = 4e-5
betas = [0.9, 0.99]
weight_decay = 0.01
eps = 1e-8</code></pre>

Files with SHA256: d7b5c187bd701455167161c7c4576e85d780402318b853766104542a7a71ec5d