WorldFoundry
English中文

World models · camera-controlled-video · image-to-video

ATI-Wan2.1-14B

Camera-Control Video

Only ATI's official inference delta is vendored; the unchanged Wan2.1 implementation is reused from worldfoundry/base_models.

IntegratedUnified environmentPinned sourceati-wan21-14b

Build your run

Choose a recorded variant, then copy the exact setup or inspection command.

worldfoundry.pipeline
Taskcamera-controlled-video
Environmentworldfoundry-unified-cu128
DeviceCUDA 12.8
worldfoundry-eval evaluate \
  --mode model \
  --model-id ati-wan21-14b \
  --model-runner worldfoundry:pipeline \
  --model-manifest-dir worldfoundry/data/models/catalog \
  --requests-path tmp/requests.jsonl \
  --output-dir tmp/model_eval/ati-wan21-14b \
  --metric artifact_count \
  --json
01

Compatibility & versions

Manifest-backed recipe

Integration
integrated
Runner evidence
pending
Environment
worldfoundry-unified-cu128
Python
3.11
CUDA
CUDA 12.8
PyTorch
torch
Source revision
1a002caf7bb55cfb016dcc670c357bd803af3a0d
Checkpoint revision
a92e2fac1f89b68f0c251942888dbdc0c15bcc4e
Runtime profile
ati-wan21-14b
Pipeline binding
ati-wan21-14b
Runner
worldfoundry.pipeline
Pipeline target
worldfoundry.pipelines.ati.pipeline_ati:ATIPipeline
Backend stage
independent_in_tree_runtime
Runtime status
checkpoint_assets_pending_not_runnable_until_staged
Driver status
compatible_unified_default
Environment kind
Unified environment
02

Install environment

The environment resolver reads the recorded profile and chooses the unified or dedicated environment shown here.

bash scripts/setup/model_env_install.sh --model ati-wan21-14b
Package constraints 4
  • torch
  • torchvision
  • torchaudio
  • xfuser
Conda packages 5
  • python=3.11
  • pip
  • ffmpeg
  • setuptools
  • wheel
03

Checkpoints & assets

Run the local check before allocating compute. Gated, private, and license fields below come directly from the checkpoint manifest.

worldfoundry-eval zoo model-download --model-id ati-wan21-14b --check-local --json
bytedance-research/ATI
Revision
—
License
—
Gated
—
Private
—
Wan-AI/Wan2.1-I2V-14B-480P
Revision
—
License
—
Gated
—
Private
—
bytedance-research/ATI
Revision
a92e2fac1f89b68f0c251942888dbdc0c15bcc4e
License
—
Gated
—
Private
—
Wan-AI/Wan2.1-I2V-14B-480P
Revision
6b73f84e66371cdfe870c72acd6826e1d61cf279
License
—
Gated
—
Private
—
04

Launch & outputs

The generated command uses the shared evaluation boundary and writes durable result manifests and artifacts.

worldfoundry-eval evaluate \
  --mode model \
  --model-id ati-wan21-14b \
  --model-runner worldfoundry:pipeline \
  --model-manifest-dir worldfoundry/data/models/catalog \
  --requests-path tmp/requests.jsonl \
  --output-dir tmp/model_eval/ati-wan21-14b \
  --metric artifact_count \
  --json

Input contract

FieldRecorded contract
promptRequired
imageRequired
dense_point_tracksRequired

Artifact contract

Artifact kindFilename / path
generated_videoati-wan21-14b.mp4
05

Evidence & sources

Catalog integration, native-demo parity, and runner parity are independent records.

Integration
integrated
Runner evidence
pending
Native demo evidence
Not recorded
Validation imports
torch, torchvision, numpy, transformers, diffusers, xfuser, worldfoundry
Recipe provenance
worldfoundry/data/models/catalog/world_models/ati-wan21-14b.yamlGitHub1a002caf7bb55cfb016dcc670c357bd803af3a0dHugging Facehttps://huggingface.co/bytedance-research/ATISourcehttps://github.com/Wan-Video/Wan2.1