Cosmos-Predict2-14B-Video2World
Cosmos-Predict2-14B-Video2World is a 14 billion parameter image-generation model from nvidia, released April 25, 2025. Cosmos-Predict2-14B-Video2World is an open-weights image model with roughly 14 billion parameters.
by nvidia · 14B parameters
Best for
Ways to use Cosmos-Predict2-14B-Video2World in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your nvidia API key. osFoundry discovers Cosmos-Predict2-14B-Video2World automatically — assign it to a Maestro role (router, direct, orchestrator, or fallback) in the Pipeline tab and it is live in every chat. Your key, your provider account — no token markup.
Deploy a dedicated endpoint
Cosmos-Predict2-14B-Video2World is open-weights — run it locally for free, or deploy a dedicated GPU endpoint in your workspace for reserved capacity with no rate limits.
Use it in a Room App
Room Apps declare AI features in their manifest, then call them with invokeAI:
import { invokeAI } from '@osfoundry/app-sdk'
// 'summarize' is an AI feature declared in your app manifest.
const result = await invokeAI('summarize', userText)
Call it from your own apps
Once a model is wired into your workspace you can host it as an API and reach it from your own services, scripts, or CI — outside osFoundry.
What hardware can run Cosmos-Predict2-14B-Video2World
Cosmos-Predict2-14B-Video2World runs on a single 16GB consumer GPU (~9 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~34 GB).
Cosmos-Predict2-14B-Video2World vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Cosmos-Predict2-14B-Video2World
Is Cosmos-Predict2-14B-Video2World free to use?
Cosmos-Predict2-14B-Video2World is free to run locally on your own hardware. Hosted access through osFoundry is metered (input Free (local), output Free (local)). You can switch between local and hosted at any time.
Can I use Cosmos-Predict2-14B-Video2World commercially?
Commercial use is allowed with conditions. Licence terms not specified — verify the upstream model card before commercial use. Check upstream documentation.
How much VRAM does Cosmos-Predict2-14B-Video2World need?
Approximately 9 GB at Q4 quantisation, or 34 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run Cosmos-Predict2-14B-Video2World locally?
Yes. Cosmos-Predict2-14B-Video2World is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Cosmos-Predict2-14B-Video2World best at?
Cosmos-Predict2-14B-Video2World is well-suited to image to video.
How do I use Cosmos-Predict2-14B-Video2World in osFoundry?
Paste your nvidia API key in the key dialog (or deploy the open weights for self-hostable models), assign Cosmos-Predict2-14B-Video2World to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by nvidia on April 25, 2025. Source: https://huggingface.co/nvidia/Cosmos-Predict2-14B-Video2World