Trigger Word: <code>Genshin_TCG</code> Model: Wan 2.1 t2i 14B All examples are generated with 1.0 LoRA strength and CFG=6 For inference used <a target="_blank" rel="ugc" href="https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/main">Kijai's workflows</a>For prompt, I recommend using the following structure:<pre><code>"Genshin_TCG medium shot" + character description (appearance, pose, clothing) + key object (weapon/artifact) + background + dynamic elements. Special attention is paid to color contrasts (dark armor vs glowing accents) and mystical atmosphere (starry sky, magical particles).</code></pre>If you want to add a gold frame like in the TGC cards, add this to the end of the prompt:<pre><code>The frame is adorned with a golden border featuring intricate designs, including star-like emblems in each corner and delicate patterns along the edges, creating an elegant and polished visual effect.</code></pre>The previous version for Wan 1.3B can be found here <a target="_blank" rel="ugc" href="https://civarchive.com/models/1728768/genshin-tcg-style-wan-13b">https://civarchive.com/models/1728768/genshin-tcg-style-wan-13b</a><h3 id="training-details-q39x2v19h">Training Details</h3>It is much easier to train the 14B version than the 1.3B. The movements are smooth, there are almost no artifacts during generation. For the training, a dataset of 54 short videos with cards from the Genshin Genius Invocation TCG card game was used. Since I used <a target="_blank" rel="ugc" href="https://github.com/tdrussell/diffusion-pipe">diffusion pipe</a> for training, I'll just post the toml files.For dataset:<pre><code>resolutions = [[514, 304]]
enable_ar_bucket = true
min_ar = 0.5
max_ar = 2.0
num_ar_buckets = 7
frame_buckets = [1, 32, 33]

[[directory]]
path = "/home/user/Genshin_TCG_dataset/videos/304_514"
num_repeats = 5
resolutions = [[514, 304]]

[[directory]]
path = "/home/user/Genshin_TCG_dataset/videos/368_620"
num_repeats = 5
resolutions = [[620, 368]]

[[directory]]
path = "/home/user/Genshin_TCG_dataset/videos/492_828"
num_repeats = 5
resolutions = [[828, 492]]</code></pre>For train:<pre><code>output_dir = "/home/user/Genshin_TCG/14B"
dataset = "/home/user/config/dataset/dataset_v001.toml"

epochs = 80
micro_batch_size_per_gpu = 1
pipeline_stages = 1
gradient_accumulation_steps = 1
gradient_clipping = 1
warmup_steps = 10
eval_every_n_epochs = 1
eval_before_first_step = true
eval_micro_batch_size_per_gpu = 1
eval_gradient_accumulation_steps = 1
save_every_n_epochs = 1
activation_checkpointing = 'unsloth'
partition_method = "parameters"
save_dtype = "bfloat16"
caching_batch_size = 1
steps_per_print = 10
video_clip_mode = "single_beginning"
blocks_to_swap = 32

[model]
type = "wan"
ckpt_path = "/home/user/Wan2.1-T2V-14B"
dtype = "bfloat16"
transformer_dtype = "float8"
timestep_sample_method = "logit_normal"

[adapter]
type = "lora"
rank = 64
dtype = "bfloat16"

[optimizer]
type = 'AdamW8bitKahan'
lr = 5e-5
betas = [0.9, 0.99]
weight_decay = 0.01
stabilize = false</code></pre>

Genshin TCG Style [Wan 14B]

genshin_tcg_r064_5e-05_wan14B.safetensors

Genshin TCG Style 

v1.0

Genshin TCG Style

wan2.1-t2v-14B

Trigger Word: Genshin_TCG Model: Wan 2.1 t2i 1.3B All examples are generated with 1.0 LoRA strength an CFG=6 For inference used <a href="https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/main" rel="ugc noopener noreferrer nofollow" target="_blank">Kijai's workflows</a>For promt, I recommend using the following structure:"Genshin_TCG medium shot" + character description (appearance, pose, clothing) + key object (weapon/artifact) + background + dynamic elements. Special attention is paid to color contrasts (dark armor vs glowing accents) and mystical atmosphere (starry sky, magical particles).If you want to add a gold frame like in the TGC cards, add this to the end of the promt:The frame is adorned with a golden border featuring intricate designs, including star-like emblems in each corner and delicate patterns along the edges, creating an elegant and polished visual effect.The previous version for Wan 1.3B can be found here <a href="https://civitai.com/models/1728768/genshin-tcg-style-wan-13b" rel="ugc noopener noreferrer nofollow" target="_blank">https://civitai.com/models/1728768/genshin-tcg-style-wan-13b</a>Training DetailsIt is much easier to train the 14B version than the 1.3B. The movements are smooth, there are almost no artifacts during generation. For the training, a dataset of 54 short videos with cards from the Genshin Genius Invocation TCG card game was used. Since I used <a href="https://github.com/tdrussell/diffusion-pipe" rel="ugc noopener noreferrer nofollow" target="_blank">diffusion pipe</a> for training, I'll just post the toml files.For dataset:resolutions = [[514, 304]] enable_ar_bucket = true min_ar = 0.5 max_ar = 2.0 num_ar_buckets = 7 frame_buckets = [1, 32, 33] [[directory]] path = "/home/user/Genshin_TCG_dataset/videos/304_514" num_repeats = 5 resolutions = [[514, 304]] [[directory]] path = "/home/user/Genshin_TCG_dataset/videos/368_620" num_repeats = 5 resolutions = [[620, 368]] [[directory]] path = "/home/user/Genshin_TCG_dataset/videos/492_828" num_repeats = 5 resolutions = [[828, 492]]For train:output_dir = "/home/user/Genshin_TCG/14B" dataset = "/home/user/config/dataset/dataset_v001.toml" epochs = 80 micro_batch_size_per_gpu = 1 pipeline_stages = 1 gradient_accumulation_steps = 1 gradient_clipping = 1 warmup_steps = 10 eval_every_n_epochs = 1 eval_before_first_step = true eval_micro_batch_size_per_gpu = 1 eval_gradient_accumulation_steps = 1 save_every_n_epochs = 1 activation_checkpointing = 'unsloth' partition_method = "parameters" save_dtype = "bfloat16" caching_batch_size = 1 steps_per_print = 10 video_clip_mode = "single_beginning" blocks_to_swap = 32 [model] type = "wan" ckpt_path = "/home/user/Wan2.1-T2V-14B" dtype = "bfloat16" transformer_dtype = "float8" timestep_sample_method = "logit_normal" [adapter] type = "lora" rank = 64 dtype = "bfloat16" [optimizer] type = 'AdamW8bitKahan' lr = 5e-5 betas = [0.9, 0.99] weight_decay = 0.01 stabilize = false

Files with SHA256: ab345301a23b484d4a64614d0709387034917024ddf0a40b7b36977d95e0b48f