Qwen3.5-397B-A17B-MLX-4.1bit
inferencerlabs's Qwen3.5-397B-A17B-MLX-4.1bit packs 397 billion parameters into a chat model. Qwen3.5-397B-A17B-MLX-4.1bit is an open-weights chat model with roughly 397 billion parameters.
by inferencerlabs · 397B parameters
Best for
Ways to use Qwen3.5-397B-A17B-MLX-4.1bit in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your inferencerlabs API key. osFoundry discovers Qwen3.5-397B-A17B-MLX-4.1bit 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
Qwen3.5-397B-A17B-MLX-4.1bit 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 Qwen3.5-397B-A17B-MLX-4.1bit
Qwen3.5-397B-A17B-MLX-4.1bit runs on a multi-GPU setup or H200 141GB at Q4 (~239 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~953 GB).
Qwen3.5-397B-A17B-MLX-4.1bit vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about Qwen3.5-397B-A17B-MLX-4.1bit
Is Qwen3.5-397B-A17B-MLX-4.1bit free to use?
Qwen3.5-397B-A17B-MLX-4.1bit 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 Qwen3.5-397B-A17B-MLX-4.1bit 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 Qwen3.5-397B-A17B-MLX-4.1bit need?
Approximately 239 GB at Q4 quantisation, or 953 GB at full FP16 precision. Requires multi-GPU at higher quantisation.
Can I run Qwen3.5-397B-A17B-MLX-4.1bit locally?
Yes. Qwen3.5-397B-A17B-MLX-4.1bit is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is Qwen3.5-397B-A17B-MLX-4.1bit best at?
Qwen3.5-397B-A17B-MLX-4.1bit is well-suited to text generation.
How do I use Qwen3.5-397B-A17B-MLX-4.1bit in osFoundry?
Paste your inferencerlabs API key in the key dialog (or deploy the open weights for self-hostable models), assign Qwen3.5-397B-A17B-MLX-4.1bit to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by inferencerlabs on February 26, 2026. Source: https://huggingface.co/inferencerlabs/Qwen3.5-397B-A17B-MLX-4.1bit