# nvidia-cosmos/cosmos-predict2.5

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

Repository: https://github.com/nvidia-cosmos/cosmos-predict2.5
Canonical: https://ross.abutalabs.com/products/cosmos-predict25
Homepage: https://research.nvidia.com/labs/dir/cosmos-predict2.5
Language: Python
License: Apache-2.0
License Family: permissive
Topics: foundational-models, video-generation, world-models
Last push: 2026-06-08T02:14:37+00:00

## Health v2 (maintenance only)
Score: 72/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 86, release rhythm 82, longevity 24
- inputs: {"age_days": 342, "days_push": 87, "days_rel": 121, "gap_med": 12.0, "n_releases_24m": 11}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1355, forks 192 (observed 2026-08-28T04:04:29.184628+00:00)

## What it is
NVIDIA Cosmos-Predict2.5 is a family of world foundation models (WFMs) that generate video predictions of future world states for physical AI applications like autonomous vehicles and robotics. The repository provides inference and post-training code, though it is now superseded by Cosmos 3 and receives only limited maintenance.

## Use cases
- generate future-state video predictions for robot training
- post-train a world model on custom robotics data
- synthesize training video for autonomous vehicle simulation
- action-conditioned video generation for embodied agents
- distill a large video world model for faster inference
- build physical AI data pipelines with synthetic video

## When to choose
- you need video-based world prediction models for robotics or AV research
- you want to post-train or distill Cosmos Predict2.5 checkpoints
- you need reproducible recipes tied to the Predict2.5 model family

## When to avoid
- you want the latest capabilities - migrate to NVIDIA Cosmos 3
- you need a lightweight video generation tool without GPU infrastructure
- you expect active development or new features in this repo

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, video-processing, llm-training, simulation
- domain: artificial-intelligence, machine-learning, deep-learning, robotics, simulation
- platform: python
- tags: world-foundation-models, video-generation, physical-ai, post-training, diffusion-models, autonomous-vehicles, robotics-simulation, nvidia, video, linux, gpu, docker

## Member repositories
- nvidia-cosmos/cosmos-predict2.5 (main) score 72

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.184628+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:41:58.731508+00:00, confidence not recorded.
  - readme: https://github.com/nvidia-cosmos/cosmos-predict2.5 (fetched 2026-08-28T04:04:29.184628+00:00, sha ed0acc6d7710)
  - homepage: https://research.nvidia.com/labs/dir/cosmos-predict2.5 (fetched 2026-08-29T12:00:18.626670+00:00, sha a6d1f4ad1566)
- Data as of 2026-08-30T08:39:29.467469+00:00.