cogito-v2-preview-deepseek-671B-MoE-4bit
Released by mlx-community in 2025, cogito-v2-preview-deepseek-671B-MoE-4bit is a 671 billion parameter chat model. cogito-v2-preview-deepseek-671B-MoE-4bit is an open-weights chat model with roughly 671 billion parameters.
by mlx-community · 671B parameters
Best for
Ways to use cogito-v2-preview-deepseek-671B-MoE-4bit in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your mlx-community API key. osFoundry discovers cogito-v2-preview-deepseek-671B-MoE-4bit 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
cogito-v2-preview-deepseek-671B-MoE-4bit 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 cogito-v2-preview-deepseek-671B-MoE-4bit
cogito-v2-preview-deepseek-671B-MoE-4bit runs on a multi-GPU setup or H200 141GB at Q4 (~403 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~1611 GB).
cogito-v2-preview-deepseek-671B-MoE-4bit vs similar models
Licence
Unspecified — Licence terms not specified — verify the upstream model card before commercial use.
Check upstream documentation.
Frequently asked about cogito-v2-preview-deepseek-671B-MoE-4bit
Is cogito-v2-preview-deepseek-671B-MoE-4bit free to use?
cogito-v2-preview-deepseek-671B-MoE-4bit 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 cogito-v2-preview-deepseek-671B-MoE-4bit 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 cogito-v2-preview-deepseek-671B-MoE-4bit need?
Approximately 403 GB at Q4 quantisation, or 1611 GB at full FP16 precision. Requires multi-GPU at higher quantisation.
Can I run cogito-v2-preview-deepseek-671B-MoE-4bit locally?
Yes. cogito-v2-preview-deepseek-671B-MoE-4bit is open-weights and runs locally on a workstation GPU. osFoundry's local runtime handles model loading, quantisation, and routing.
What is cogito-v2-preview-deepseek-671B-MoE-4bit best at?
cogito-v2-preview-deepseek-671B-MoE-4bit is well-suited to text generation.
How do I use cogito-v2-preview-deepseek-671B-MoE-4bit in osFoundry?
Paste your mlx-community API key in the key dialog (or deploy the open weights for self-hostable models), assign cogito-v2-preview-deepseek-671B-MoE-4bit to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by mlx-community on August 2, 2025. Source: https://huggingface.co/mlx-community/cogito-v2-preview-deepseek-671B-MoE-4bit