vila-u-7b-256
vila-u-7b-256 is a 7 billion parameter chat model from mit-han-lab, released October 21, 2024. vila-u-7b-256 is an open-weights chat model with roughly 7 billion parameters.
by mit-han-lab · 7B parameters
Best for
Ways to use vila-u-7b-256 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your mit-han-lab API key. osFoundry discovers vila-u-7b-256 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
vila-u-7b-256 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 vila-u-7b-256
vila-u-7b-256 runs on a single 16GB consumer GPU (~5 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~17 GB).
vila-u-7b-256 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about vila-u-7b-256
Is vila-u-7b-256 free to use?
vila-u-7b-256 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 vila-u-7b-256 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 vila-u-7b-256 need?
Approximately 5 GB at Q4 quantisation, or 17 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run vila-u-7b-256 locally?
Yes. vila-u-7b-256 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is vila-u-7b-256 best at?
vila-u-7b-256 is well-suited to any to any.
How do I use vila-u-7b-256 in osFoundry?
Paste your mit-han-lab API key in the key dialog (or deploy the open weights for self-hostable models), assign vila-u-7b-256 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by mit-han-lab on October 21, 2024. Source: https://huggingface.co/mit-han-lab/vila-u-7b-256