ministral-3-3b-instruct-2512-null-space-abliterated
Built by jwest33, ministral-3-3b-instruct-2512-null-space-abliterated is a 3 billion parameter image-generation model. ministral-3-3b-instruct-2512-null-space-abliterated is an open-weights image model with roughly 3 billion parameters.
by jwest33 · 3B parameters
Best for
Ways to use ministral-3-3b-instruct-2512-null-space-abliterated in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your jwest33 API key. osFoundry discovers ministral-3-3b-instruct-2512-null-space-abliterated 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
ministral-3-3b-instruct-2512-null-space-abliterated 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 ministral-3-3b-instruct-2512-null-space-abliterated
ministral-3-3b-instruct-2512-null-space-abliterated 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 (~8 GB).
ministral-3-3b-instruct-2512-null-space-abliterated vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about ministral-3-3b-instruct-2512-null-space-abliterated
Is ministral-3-3b-instruct-2512-null-space-abliterated free to use?
ministral-3-3b-instruct-2512-null-space-abliterated 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 ministral-3-3b-instruct-2512-null-space-abliterated 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 ministral-3-3b-instruct-2512-null-space-abliterated need?
Approximately 2 GB at Q4 quantisation, or 8 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run ministral-3-3b-instruct-2512-null-space-abliterated locally?
Yes. ministral-3-3b-instruct-2512-null-space-abliterated is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is ministral-3-3b-instruct-2512-null-space-abliterated best at?
ministral-3-3b-instruct-2512-null-space-abliterated is well-suited to image text to text.
How do I use ministral-3-3b-instruct-2512-null-space-abliterated in osFoundry?
Paste your jwest33 API key in the key dialog (or deploy the open weights for self-hostable models), assign ministral-3-3b-instruct-2512-null-space-abliterated to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by jwest33 on January 20, 2026. Source: https://huggingface.co/jwest33/ministral-3-3b-instruct-2512-null-space-abliterated