Models#

FlashDreams runs a growing family of world and video models (text-to-video, image-to-video, camera-controlled, and super-resolution), all through one consistent command line and Python interface. Browse the models below, pick the one that fits what you want to make, and follow its card through to the full method.

Available models#

The models come in three flavors. Streaming and autoregressive generation methods build a video step by step and stay fast once warmed up, aiming for sub-second latency per step; bidirectional methods produce a clip in a single pass and serve as the quality reference for their streaming counterparts; and super-resolution methods upscale existing frames in chunks, so their latency scales with output resolution rather than step count. Each card links to that method’s page, where you’ll find the exact command to run it, the checkpoint it uses, and the settings you can tune.

Streaming and autoregressive generation

OmniDreams

Interactive world simulator for autonomous vehicles.

NVIDIA OmniDreams
Self-Forcing

Autoregressive text-to-video based on Wan 2.1.

Self-Forcing
Causal-Forcing

Autoregressive text/image-to-video based on Wan 2.1.

Causal-Forcing
Causal Wan 2.2

Autoregressive text-to-video based on Wan 2.2 from FastVideo.

Causal Wan2.2
LingBot-World

Camera-controllable image-to-video world model.

LingBot-World
Waypoint 1.5

Interactive image-established world model controlled by keyboard and mouse.

Waypoint 1.5
HY-WorldPlay

Action- and camera-controllable image-to-video world model.

HY-WorldPlay
SANA-WM_streaming
SANA-WM streaming FlashDreams sample clip.

Chunk-causal camera-controlled world model with streaming Stage-1, refiner, and VAE paths.

SANA-WM_streaming

Bidirectional Video Generation

Wan 2.1

Bidirectional video generation model that supports both text-to-video and image-to-video.

Wan2.1
Cosmos-Predict2.5

Bidirectional Cosmos-Predict2 reference implementations (T2V / I2V, 2B).

Cosmos-Predict2.5
SANA-WM_bidirectional
SANA-WM bidirectional FlashDreams sample clip.

Bidirectional camera-controlled world model (Stage-1 DiT + LTX-2 refiner, 2.6B).

SANA-WM_bidirectional

Super-resolution

FlashVSR

Streaming video super-resolution.

FlashVSR

Running a model yourself#

uv run flashdreams-run <MODEL_SLUG> --help

Examples:

uv run --project integrations_v2/self_forcing flashdreams-run-v2 \
    t2v-self-forcing-wan2.1-t2v-1.3b --output-path artifacts/t2v-self-forcing-wan2.1-t2v-1.3b.mp4 -- \
    --prompt "A cat surfing." --total-blocks 7
uv run --no-sync flashdreams-run-v2 cam2v-lingbot \
    --mode mp4 --output-path outputs/lingbot.mp4 -- \
    --example-data --total-blocks 21

Adding your own model#

See Add a new method for model integration and registration guidance.