ModelListedExternal

LongLive 1.0 Stage 1 Diffusers

Diffusers-style LongLive stage-1 checkpoint converted from:

Sourcehuggingface.co/mignonjia/LongLive-1.0-Stage-1-Diffusers/tree/e89b1b177ee0bc8097d3e333f77ee5c89d03147c

KindOtherVersionRevision e89b1b1Cgrade

Model source

Kind: Other
File size: 484 MB–5.29 GB
Version: vRevision e89b1b1
Updated: Oct 6, 2026
Downloads: 4
Source: Hugging Face

Source description

LongLive 1.0 Stage 1 Diffusers

Diffusers-style LongLive stage-1 checkpoint converted from:

/mnt/weka/home/hao.zhang/mhuo/FastVideo/LongLive/longlive_models/models/longlive_base.pt

The model uses the causal Wan 2.1 T2V 1.3B layout. The LongLive weights replace the transformer checkpoint; scheduler, tokenizer, text encoder, and VAE follow wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers.

Recommended runtime kwargs:

model_kwargs:
  timestep_shift: 5.0
  local_attn_size: 12
  sink_size: 3

FastVideo training config equivalent:

models:
  student:
    _target_: fastvideo.train.models.wan.WanCausalModel
    init_from: mignonjia/LongLive-1.0-Stage-1-Diffusers
    trainable: true

pipeline:
  flow_shift: 5
  dit_config:
    local_attn_size: 12
    sink_size: 3

Sources

1 source

Hugging FacePrimaryVersion e89b1b177ee0bc8097d3e333f77ee5c89d03147c

huggingface.co/mignonjia/LongLive-1.0-Stage-1-Diffusers/tree/e89b1b177ee0bc8097d3e333f77ee5c89d03147c

Verified Aug 24

Model artifacts

7 artifacts

text_encoder/model-00001-of-00005.safetensors

safetensors · 4.63 GB · SHA-256 c0ef3a140898…9f99 · Hugging Face

Download

text_encoder/model-00002-of-00005.safetensors

safetensors · 4.56 GB · SHA-256 481c7b2b3977…29ad · Hugging Face

Download

text_encoder/model-00003-of-00005.safetensors

safetensors · 4.63 GB · SHA-256 f93148bcc040…35c3 · Hugging Face

Download

text_encoder/model-00004-of-00005.safetensors

safetensors · 4.66 GB · SHA-256 a451792c739c…57c9 · Hugging Face

Download

text_encoder/model-00005-of-00005.safetensors

safetensors · 2.69 GB · SHA-256 7e76e18d2245…4eea · Hugging Face

Download

transformer/diffusion_pytorch_model.safetensors

safetensors · 5.29 GB · SHA-256 264eaae28efd…393d · Hugging Face

Download

vae/diffusion_pytorch_model.safetensors

safetensors · 484 MB · SHA-256 d6e524b3fffe…e793 · Hugging Face

Download

Source excerpts

2 excerpts

ReadmeAug 24huggingface.co/mignonjia/LongLive-1.0-Stage-1-Diffusers

--- license: apache-2.0 library_name: diffusers pipeline_tag: text-to-video tags: - longlive - wan - text-to-video - fastvideo --- # LongLive 1.0 Stage 1 Diffusers Diffusers-style LongLive stage-1 checkpoint converted from: `/mnt/weka/home/hao.zhang/mhuo/FastVideo/LongLive/longlive_models/models/longlive_base.pt` The model uses the causal Wan 2.1 T2V 1.3B layout. The LongLive weights replace the transformer checkpoint; scheduler, tokenizer, text encoder, and VAE follow `wlsaidhi/SFWan2.1-T2V-1.3B-Diffusers`. Recommended runtime kwargs: ```yaml model_kwargs: timestep_shift: 5.0 local_attn_size: 12 sink_size: 3 ``` FastVideo training config equivalent: ```yaml models: student: _target_: fastvideo.train.models.wan.WanCausalModel init_from: mignonjia/LongLive-1.0-Stage-1-Diffusers trainable: true pipeline: flow_shift: 5 dit_config: local_attn_size: 12 sink_size: 3 ```

NoteAug 24huggingface.co/mignonjia/LongLive-1.0-Stage-1-Diffusers

Source context: 4 downloads · 0 likes · Pipeline text-to-video · Library diffusers · Repo mignonjia/LongLive-1.0-Stage-1-Diffusers