LightX2V supports distributed parallel inference, enabling the utilization of multiple GPUs for inference. The DiT component supports two parallel attention mechanisms: Ulysses and Ring, while also supporting Cfg parallel inference. Parallel inference significantly reduces inference time and alleviates memory overhead on each GPU.
Configuration method:
"parallel": {
"seq_p_size": 4,
"seq_p_attn_type": "ulysses"
}Configuration method:
"parallel": {
"seq_p_size": 4,
"seq_p_attn_type": "ring"
}Configuration method:
"parallel": {
"cfg_p_size": 2
}Configuration method:
"parallel": {
"seq_p_size": 4,
"seq_p_attn_type": "ulysses",
"cfg_p_size": 2
}Parallel inference configuration files are available here
By specifying --config_json to a specific config file, you can test parallel inference.
Here are some run scripts for your use.