QwQ-32B-MLX-8bit
lmstudio-community's QwQ-32B-MLX-8bit packs 32 billion parameters into a chat model. QwQ-32B-MLX-8bit is an open-weights chat model with roughly 32 billion parameters.
by lmstudio-community · 32B parameters
Best for
Ways to use QwQ-32B-MLX-8bit in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your lmstudio-community API key. osFoundry discovers QwQ-32B-MLX-8bit 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
QwQ-32B-MLX-8bit 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 QwQ-32B-MLX-8bit
QwQ-32B-MLX-8bit runs on a 24GB consumer or workstation GPU (~20 GB VRAM with KV-cache headroom). Full-precision inference fits on a single H100 80GB at FP16 precision (~77 GB).
QwQ-32B-MLX-8bit vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about QwQ-32B-MLX-8bit
Is QwQ-32B-MLX-8bit free to use?
QwQ-32B-MLX-8bit 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 QwQ-32B-MLX-8bit 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 QwQ-32B-MLX-8bit need?
Approximately 20 GB at Q4 quantisation, or 77 GB at full FP16 precision. Fits on a single 24GB consumer GPU.
Can I run QwQ-32B-MLX-8bit locally?
Yes. QwQ-32B-MLX-8bit is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is QwQ-32B-MLX-8bit best at?
QwQ-32B-MLX-8bit is well-suited to text generation.
How do I use QwQ-32B-MLX-8bit in osFoundry?
Paste your lmstudio-community API key in the key dialog (or deploy the open weights for self-hostable models), assign QwQ-32B-MLX-8bit to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by lmstudio-community on March 5, 2025. Source: https://huggingface.co/lmstudio-community/QwQ-32B-MLX-8bit