Anima

Anima is a Diffusion Transformer (DiT) model based on the MiniTrainDIT design with Rectified Flow training. It uses a Qwen3-0.6B text encoder, an LLM Adapter bridge, and a Qwen-Image VAE with a 16-channel, 8× spatial downscale latent space.

Anima support was added in sd-scripts v0.10.1. Model weights are available from the circlestone-labs/Anima repository on Hugging Face.

Supported training methods

Method	Script	Supported
LoRA	`anima_train_network.py`	Yes
Full fine-tuning	`anima_train.py`	Yes
LoHa / LoKr	`anima_train_network.py`	Yes (experimental)
Textual Inversion	—	No
ControlNet	—	No

Required model files

You need four components before training:

Component	Description	Source
Anima DiT	Base DiT model `.safetensors`	circlestone-labs/Anima
Qwen3-0.6B	Text encoder (HuggingFace dir or `.safetensors`)	Qwen3-0.6B
Qwen-Image VAE	VAE model `.safetensors` or `.pth`	circlestone-labs/Anima
LLM Adapter	6-layer Transformer bridge (optional, loaded from DiT if bundled)	Bundled in DiT file

The LLM Adapter is typically bundled inside the DiT model file. If the key llm_adapter.out_proj.weight is present in the DiT weights, you do not need to specify --llm_adapter_path separately.

Training command

accelerate launch --num_cpu_threads_per_process 1 anima_train_network.py \
  --pretrained_model_name_or_path="/path/to/anima-dit.safetensors" \
  --qwen3="/path/to/Qwen3-0.6B" \
  --vae="/path/to/qwen-image-vae.safetensors" \
  --dataset_config="my_anima_dataset_config.toml" \
  --output_dir="./output" \
  --output_name="my_anima_lora" \
  --save_model_as=safetensors \
  --network_module=networks.lora_anima \
  --network_dim=8 \
  --learning_rate=1e-4 \
  --optimizer_type="AdamW8bit" \
  --lr_scheduler="constant" \
  --timestep_sampling="sigmoid" \
  --discrete_flow_shift=1.0 \
  --max_train_epochs=10 \
  --save_every_n_epochs=1 \
  --mixed_precision="bf16" \
  --gradient_checkpointing \
  --cache_latents \
  --cache_text_encoder_outputs \
  --vae_chunk_size=64 \
  --vae_disable_cache

If training loss becomes NaN, verify that your PyTorch version is 2.5 or higher.

Key Anima-specific arguments

string

required

Path to the Anima DiT model .safetensors file. ComfyUI format with a net. key prefix is supported.

string

required

Path to the Qwen3-0.6B text encoder. Can be a HuggingFace model directory or a single .safetensors file. The text encoder is always frozen during training.

string

required

Path to the Qwen-Image VAE .safetensors or .pth file. The architecture is fixed: dim=96, z_dim=16.

string

Path to a separate LLM adapter weights file. If omitted, the adapter is loaded from the DiT file when the key llm_adapter.out_proj.weight exists.

string

default:"sigmoid"

Timestep sampling method. Options: sigma, uniform, sigmoid, shift, flux_shift. Same options as FLUX training.

number

default:"1.0"

Shift for the timestep distribution. Used when --timestep_sampling=shift.

integer

Process the VAE in chunks of this size to reduce memory usage. Recommended: 64.

boolean

Disables the VAE cache to reduce memory usage. Use alongside --vae_chunk_size.

string

default:"torch"

Attention implementation. Options: torch, xformers, flash, sageattn. Note: sageattn is inference-only and cannot be used for training.

Network module

Use networks.lora_anima as the --network_module:

--network_module=networks.lora_anima

LoHa and LoKr are also supported for Anima — see the LoHa / LoKr page for details.

Converting to ComfyUI format

After training, convert the LoRA to ComfyUI format using:

python networks/convert_anima_lora_to_comfy.py \
  --input "/path/to/my_anima_lora.safetensors" \
  --output "/path/to/my_anima_lora_comfy.safetensors"

Supported Models

Network Modules

Supported training methods

Required model files

Training command

Key Anima-specific arguments

Network module

Converting to ComfyUI format

​Supported training methods

​Required model files

​Training command

​Key Anima-specific arguments

​Network module

​Converting to ComfyUI format

Supported training methods

Required model files

Training command

Key Anima-specific arguments

Network module

Converting to ComfyUI format