cogito-v2-preview-llama-109B-MoE-mlx-2Bit
Built by k8smee, cogito-v2-preview-llama-109B-MoE-mlx-2Bit is a 109 billion parameter image-generation model. cogito-v2-preview-llama-109B-MoE-mlx-2Bit is an open-weights image model with roughly 109 billion parameters.
by k8smee · 109B parameters
Best for
Ways to use cogito-v2-preview-llama-109B-MoE-mlx-2Bit in osFoundry
Connect with your own key (BYOK)
Open the key dialog and paste your k8smee API key. osFoundry discovers cogito-v2-preview-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit
cogito-v2-preview-llama-109B-MoE-mlx-2Bit runs on a single A100 80GB or H100 80GB at Q4 quantisation (~66 GB VRAM with KV-cache headroom). Full-precision inference requires multiple H100/H200 GPUs at FP16 (~262 GB).
cogito-v2-preview-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit
Is cogito-v2-preview-llama-109B-MoE-mlx-2Bit free to use?
cogito-v2-preview-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit need?
Approximately 66 GB at Q4 quantisation, or 262 GB at full FP16 precision. Fits on a single A100/H100 80GB.
Can I run cogito-v2-preview-llama-109B-MoE-mlx-2Bit locally?
Yes. cogito-v2-preview-llama-109B-MoE-mlx-2Bit 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-llama-109B-MoE-mlx-2Bit best at?
cogito-v2-preview-llama-109B-MoE-mlx-2Bit is well-suited to image text to text.
How do I use cogito-v2-preview-llama-109B-MoE-mlx-2Bit in osFoundry?
Paste your k8smee API key in the key dialog (or deploy the open weights for self-hostable models), assign cogito-v2-preview-llama-109B-MoE-mlx-2Bit to a Maestro role in the Pipeline tab, then use it in chat, Room Apps via invokeAI, or your own apps.
Published by k8smee on March 20, 2026. Source: https://huggingface.co/k8smee/cogito-v2-preview-llama-109B-MoE-mlx-2Bit