pixtral-12b-FP8-dynamic
Released by nm-testing in 2025, pixtral-12b-FP8-dynamic is a 12 billion parameter image-generation model. pixtral-12b-FP8-dynamic is an open-weights image model with roughly 12 billion parameters.
by nm-testing · 12B parameters
Best for
Ways to use pixtral-12b-FP8-dynamic in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your nm-testing API key. osFoundry discovers pixtral-12b-FP8-dynamic 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
pixtral-12b-FP8-dynamic 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 pixtral-12b-FP8-dynamic
pixtral-12b-FP8-dynamic runs on a single 16GB consumer GPU (~8 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~29 GB).
pixtral-12b-FP8-dynamic vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about pixtral-12b-FP8-dynamic
Is pixtral-12b-FP8-dynamic free to use?
pixtral-12b-FP8-dynamic 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 pixtral-12b-FP8-dynamic 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 pixtral-12b-FP8-dynamic need?
Approximately 8 GB at Q4 quantisation, or 29 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run pixtral-12b-FP8-dynamic locally?
Yes. pixtral-12b-FP8-dynamic is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is pixtral-12b-FP8-dynamic best at?
pixtral-12b-FP8-dynamic is well-suited to image text to text.
How do I use pixtral-12b-FP8-dynamic in osFoundry?
Paste your nm-testing API key in the key dialog (or deploy the open weights for self-hostable models), assign pixtral-12b-FP8-dynamic to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by nm-testing on February 8, 2025. Source: https://huggingface.co/nm-testing/pixtral-12b-FP8-dynamic