Cosmos-Reason2-2B-W4A16-Edge2-FlashHead
Built by embedl, Cosmos-Reason2-2B-W4A16-Edge2-FlashHead is a 2 billion parameter image-generation model. Cosmos-Reason2-2B-W4A16-Edge2-FlashHead is an open-weights image model with roughly 2 billion parameters.
by embedl · 2B parameters
Best for
Ways to use Cosmos-Reason2-2B-W4A16-Edge2-FlashHead in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your embedl API key. osFoundry discovers Cosmos-Reason2-2B-W4A16-Edge2-FlashHead 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-Reason2-2B-W4A16-Edge2-FlashHead 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-Reason2-2B-W4A16-Edge2-FlashHead
Cosmos-Reason2-2B-W4A16-Edge2-FlashHead runs on a single 16GB consumer GPU (~2 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~5 GB).
Cosmos-Reason2-2B-W4A16-Edge2-FlashHead vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Cosmos-Reason2-2B-W4A16-Edge2-FlashHead
Is Cosmos-Reason2-2B-W4A16-Edge2-FlashHead free to use?
Cosmos-Reason2-2B-W4A16-Edge2-FlashHead 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-Reason2-2B-W4A16-Edge2-FlashHead 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-Reason2-2B-W4A16-Edge2-FlashHead need?
Approximately 2 GB at Q4 quantisation, or 5 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run Cosmos-Reason2-2B-W4A16-Edge2-FlashHead locally?
Yes. Cosmos-Reason2-2B-W4A16-Edge2-FlashHead is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Cosmos-Reason2-2B-W4A16-Edge2-FlashHead best at?
Cosmos-Reason2-2B-W4A16-Edge2-FlashHead is well-suited to image text to text.
How do I use Cosmos-Reason2-2B-W4A16-Edge2-FlashHead in osFoundry?
Paste your embedl API key in the key dialog (or deploy the open weights for self-hostable models), assign Cosmos-Reason2-2B-W4A16-Edge2-FlashHead to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by embedl on February 25, 2026. Source: https://huggingface.co/embedl/Cosmos-Reason2-2B-W4A16-Edge2-FlashHead