Video · text-to-video · image-to-video

LTX-Video

Lightricks

License is reported as 'other' in the official HF cards.

Integrated统一环境源码已固定ltx-video

构建运行命令

选择仓库中已记录的 variant，然后复制对应的准备、检查或运行命令。

worldfoundry.pipeline

Variant

任务image-to-video

环境worldfoundry-unified-cu128

设备CUDA 12.8

worldfoundry-eval evaluate \
  --mode model \
  --model-id ltx-video-i2v \
  --model-runner worldfoundry:pipeline \
  --model-manifest-dir worldfoundry/data/models/catalog \
  --requests-path tmp/requests.jsonl \
  --output-dir tmp/model_eval/ltx-video-i2v \
  --metric artifact_count \
  --json

兼容性与版本

由 Manifest 生成

集成状态: integrated
Runner 证据: pending
环境: worldfoundry-unified-cu128
Python: 3.11
CUDA: CUDA 12.8
PyTorch: torch
源码 revision: 4b2d053057623ddd4d0a1d3e9cd28890e9ef487f
Checkpoint revision: 8984fa25007f376c1a299016d0957a37a2f797bb

Runtime profile: ltx-video-i2v
Pipeline binding: ltx-video
Runner: worldfoundry.pipeline
Pipeline target: worldfoundry.pipelines.ltx_video.pipeline_ltx_video_i2v:LTXVideoI2VPipeline
Backend stage: official_in_tree_runtime
Runtime 状态: official_in_tree_runtime_ready_checkpoint_required
Driver 状态: compatible_unified_default
环境类型: 统一环境

安装环境

环境解析器会读取已记录 profile，并选择此处显示的统一或独立环境。

bash scripts/setup/model_env_install.sh --model ltx-video-i2v

依赖版本约束 4

torch
torchvision
torchaudio
xfuser

Conda 依赖 5

python=3.11
pip
ffmpeg
setuptools
wheel

Checkpoint 与资产

分配算力前先做本地检查；gated、private 与 license 字段直接来自 checkpoint manifest。

worldfoundry-eval zoo model-download --model-id ltx-video --check-local --json

Lightricks/LTX-Video

Revision: 8984fa25007f376c1a299016d0957a37a2f797bb
License: other
Gated: false
Private: false

Lightricks/LTX-Video-0.9.5

Revision: e58e28c39631af4d1468ee57a853764e11c1d37e
License: other
Gated: false
Private: false

Lightricks/LTX-Video-0.9.7-dev

Revision: 2101082f5eb5540770a2df43747feadb6f69b889
License: other
Gated: false
Private: false

Lightricks/LTX-Video-0.9.8-13B-distilled

Revision: 7c64400e1861cc0d7b98d570a1926d5408ec60cd
License: other
Gated: false
Private: false

运行与输出

生成的命令通过共享 evaluation 边界运行，并持久保存结果 manifest 与 artifact。

worldfoundry-eval evaluate \
  --mode model \
  --model-id ltx-video-i2v \
  --model-runner worldfoundry:pipeline \
  --model-manifest-dir worldfoundry/data/models/catalog \
  --requests-path tmp/requests.jsonl \
  --output-dir tmp/model_eval/ltx-video-i2v \
  --metric artifact_count \
  --json

输入契约

字段	已记录契约
`prompt`	Required
`image`	Required
`video`	Optional
`actions`	seed

Artifact 契约

Artifact 类型	文件名 / 路径
`generated_video`	`ltx-video-i2v.mp4`

证据与来源

Catalog 集成、原生 demo parity 与 runner parity 是三条独立记录。

集成状态: integrated
Runner 证据: pending
原生 Demo 证据: pending
验证 Imports: torch, torchvision, numpy, transformers, diffusers, xfuser, worldfoundry

配方溯源

worldfoundry/data/models/catalog/video/ltx-video.yamlGitHub4b2d053057623ddd4d0a1d3e9cd28890e9ef487f

构建运行命令

输入契约

Artifact 契约

相关配方