Cosmos Predict 2

Cosmos Predict 2 is NVIDIA's previous-generation world foundation model for image/video-to-world. Text plus a first frame or video prefix go in; a future world video comes out. This card is not Predict 2.5, not Transfer 2.5, and not Cosmos 3.

WorldFoundry assembles it natively — framework-owned denoiser, T5-11B conditioner, Rectified Flow AB2 scheduler, Wan codec, shared loader. No Diffusers pipeline and no NVIDIA runtime package. Recipe defaults are 1280×704, 93 frames, 16 fps, 35 steps. Variants are only cosmos-predict2-2b-video2world and cosmos-predict2-14b-video2world. Aliases cosmos-predict2 and cosmos2 resolve here.

The official Predict 2 project is research.nvidia.com/labs/dir/cosmos-predict2. The catalog paper URL is the 2025 Cosmos World Foundation Model Platform write-up (Predict1-era platform). That is lineage, not the Predict 2.5 paper.

What this repository actually checked: native 2B architecture 567 checkpoint tensors (no synthetic cross-attention projection), 14B 727 tensors, plus registry, conditioning masks, and AB2 numerics. GPU output parity is pending — the official Video2World weights are gated (nvidia-open-model-license) and were not in the authenticated cache used for that check.

worldfoundry-eval run \
  cosmos-predict2-2b-video2world \
  --pipeline.task-profile default \
  --json

What it is for

Stay on this card only when you need the T5-conditioned v2 recipe. New world-video work should start on Predict 2.5: Reason1 conditioner, flow-UniPC, and a verified 2B text-to-world route at the same 1280×704 / 93-frame / 35-step geometry.

Launch worldfoundry-eval run cosmos-predict2-2b-video2world. Required inputs: prompt, image, video. Artifact: cosmos-predict2-2b-video2world.mp4 (14B writes cosmos-predict2-14b-video2world.mp4). There is no control-video / VACE branch here — that is Transfer 2.5. There is no joint sound or policy head — that is Cosmos 3.

Score predicted world video on WorldModelBench and Physics-IQ. Those are catalog entry points, not ingested WorldFoundry scores.

Running in WorldFoundry

CosmosPredict2Pipeline runs in worldfoundry-unified-cu128 (Python 3.11, CUDA 12.8). Stage nvidia/Cosmos-Predict2-2B-Video2World or the 14B twin, plus google-t5/t5-11b, after accepting NVIDIA's gated terms. Predict 2 and Predict 2.5 share one checkpoint-shaped transformer implementation in-tree; version-specific architecture and sampling live in the recipes.

Related models

Open-source research release (institution not recorded)Astra

Open recipe

The Chinese University of Hong Kong / Shanghai AI LaboratoryCameraCtrl

Open recipe

NVIDIACosmos Transfer 2.5

Open recipe

University of Geneva / University of Edinburgh / Microsoft ResearchDIAMOND

Open recipe