MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 is a 172 billion parameter chat model from MJPansa, released April 15, 2026. MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 is an open-weights chat model with roughly 172 billion parameters.
by MJPansa · 172B parameters
Best for
Ways to use MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your MJPansa API key. osFoundry discovers MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 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
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 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 MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 runs on a multi-GPU setup or H200 141GB at Q4 (~104 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~413 GB).
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16
Is MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 free to use?
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 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 MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 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 MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 need?
Approximately 104 GB at Q4 quantisation, or 413 GB at full FP16 precision. Requires multi-GPU at higher quantisation.
Can I run MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 locally?
Yes. MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 best at?
MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 is well-suited to text generation.
How do I use MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 in osFoundry?
Paste your MJPansa API key in the key dialog (or deploy the open weights for self-hostable models), assign MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by MJPansa on April 15, 2026. Source: https://huggingface.co/MJPansa/MiniMax-M2.7-REAP-172B-A10B-AutoRound-W4A16