Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g
Built by ISTA-DASLab, Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g is a 24 billion parameter image-generation model. Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g is an open-weights image model with roughly 24 billion parameters.
by ISTA-DASLab · 24B parameters
Best for
Ways to use Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your ISTA-DASLab API key. osFoundry discovers Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g 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
Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g 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 Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g
Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g runs on a single 16GB consumer GPU (~15 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~58 GB).
Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g
Is Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g free to use?
Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g 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 Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g 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 Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g need?
Approximately 15 GB at Q4 quantisation, or 58 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g locally?
Yes. Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g best at?
Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g is well-suited to image text to text.
How do I use Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g in osFoundry?
Paste your ISTA-DASLab API key in the key dialog (or deploy the open weights for self-hostable models), assign Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by ISTA-DASLab on March 20, 2025. Source: https://huggingface.co/ISTA-DASLab/Mistral-Small-3.1-24B-Instruct-2503-GPTQ-4b-128g