magnum-v4-72b-W4A16-GPTQ
magnum-v4-72b-W4A16-GPTQ is a 72 billion parameter chat model from andriadze, released July 31, 2025. magnum-v4-72b-W4A16-GPTQ is an open-weights chat model with roughly 72 billion parameters.
by andriadze · 72B parameters
Best for
- complex multi-step reasoning
- agent orchestration with tool use
- long-document analysis and summarisation
Ways to use magnum-v4-72b-W4A16-GPTQ in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your andriadze API key. osFoundry discovers magnum-v4-72b-W4A16-GPTQ 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
magnum-v4-72b-W4A16-GPTQ 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 magnum-v4-72b-W4A16-GPTQ
magnum-v4-72b-W4A16-GPTQ runs on a single A100 80GB or H100 80GB at Q4 quantisation (~44 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~173 GB).
magnum-v4-72b-W4A16-GPTQ vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about magnum-v4-72b-W4A16-GPTQ
Is magnum-v4-72b-W4A16-GPTQ free to use?
magnum-v4-72b-W4A16-GPTQ 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 magnum-v4-72b-W4A16-GPTQ 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 magnum-v4-72b-W4A16-GPTQ need?
Approximately 44 GB at Q4 quantisation, or 173 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run magnum-v4-72b-W4A16-GPTQ locally?
Yes. magnum-v4-72b-W4A16-GPTQ is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is magnum-v4-72b-W4A16-GPTQ best at?
magnum-v4-72b-W4A16-GPTQ is well-suited to complex multi-step reasoning, agent orchestration with tool use, long-document analysis and summarisation.
How do I use magnum-v4-72b-W4A16-GPTQ in osFoundry?
Paste your andriadze API key in the key dialog (or deploy the open weights for self-hostable models), assign magnum-v4-72b-W4A16-GPTQ to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by andriadze on July 31, 2025. Source: https://huggingface.co/andriadze/magnum-v4-72b-W4A16-GPTQ