Devstral-2-123B-Instruct-2512
Devstral-2-123B-Instruct-2512 (mistralai, 2025) is a 123 billion parameter chat model. Devstral-2-123B-Instruct-2512 is an open-weights chat model with roughly 123 billion parameters.
by mistralai · 123B parameters
Best for
- complex multi-step reasoning
- agent orchestration with tool use
- long-document analysis and summarisation
Ways to use Devstral-2-123B-Instruct-2512 in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your mistralai API key. osFoundry discovers Devstral-2-123B-Instruct-2512 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
Devstral-2-123B-Instruct-2512 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.
Use Devstral-2-123B-Instruct-2512 via API
Devstral-2-123B-Instruct-2512 is also served by hosted API providers — use it via API (BYOK) if you would rather not manage GPUs. That page lists per-provider pricing.
What hardware can run Devstral-2-123B-Instruct-2512
Devstral-2-123B-Instruct-2512 runs on a single A100 80GB or H100 80GB at Q4 quantisation (~74 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~296 GB).
Devstral-2-123B-Instruct-2512 vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Devstral-2-123B-Instruct-2512
Is Devstral-2-123B-Instruct-2512 free to use?
Devstral-2-123B-Instruct-2512 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 Devstral-2-123B-Instruct-2512 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 Devstral-2-123B-Instruct-2512 need?
Approximately 74 GB at Q4 quantisation, or 296 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run Devstral-2-123B-Instruct-2512 locally?
Yes. Devstral-2-123B-Instruct-2512 is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Devstral-2-123B-Instruct-2512 best at?
Devstral-2-123B-Instruct-2512 is well-suited to complex multi-step reasoning, agent orchestration with tool use, long-document analysis and summarisation.
How do I use Devstral-2-123B-Instruct-2512 in osFoundry?
Paste your mistralai API key in the key dialog (or deploy the open weights for self-hostable models), assign Devstral-2-123B-Instruct-2512 to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by mistralai on November 28, 2025. Source: https://huggingface.co/mistralai/Devstral-2-123B-Instruct-2512