medgemma-27b-it-8bit
medgemma-27b-it-8bit (mlx-community, 2025) is a 27 billion parameter image-generation model. medgemma-27b-it-8bit is an open-weights image model with roughly 27 billion parameters.
by mlx-community · 27B parameters
Best for
Ways to use medgemma-27b-it-8bit in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your mlx-community API key. osFoundry discovers medgemma-27b-it-8bit 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
medgemma-27b-it-8bit 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 medgemma-27b-it-8bit
medgemma-27b-it-8bit runs on a 24GB consumer or workstation GPU (~17 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~65 GB).
medgemma-27b-it-8bit vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about medgemma-27b-it-8bit
Is medgemma-27b-it-8bit free to use?
medgemma-27b-it-8bit 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 medgemma-27b-it-8bit 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 medgemma-27b-it-8bit need?
Approximately 17 GB at Q4 quantisation, or 65 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run medgemma-27b-it-8bit locally?
Yes. medgemma-27b-it-8bit is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is medgemma-27b-it-8bit best at?
medgemma-27b-it-8bit is well-suited to image text to text.
How do I use medgemma-27b-it-8bit in osFoundry?
Paste your mlx-community API key in the key dialog (or deploy the open weights for self-hostable models), assign medgemma-27b-it-8bit to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by mlx-community on July 10, 2025. Source: https://huggingface.co/mlx-community/medgemma-27b-it-8bit