---
title: GrootTeam/Groot-280B-A16B-261023
url: https://recentmodel.com/models/GrootTeam/Groot-280B-A16B-261023/
status: just landed · new, unvetted
published: 2026-10-23T09:51:27Z
updated: 2026-10-23T11:26:00Z
source: https://huggingface.co/GrootTeam/Groot-280B-A16B-261023
---

# GrootTeam/Groot-280B-A16B-261023

Groot is a family of open-weight reasoning models for complex real-world tasks. Mixture-of-experts, 491B total, 16B active (192 experts, 6 per token). Context: 1M tokens (config). License: mit.


- **Org:** GrootTeam
- **Released:** 2026-10-23
- **Status:** just landed · new, unvetted
- **License:** mit (permissive)
- **Parameters:** 491.4B total, 16.0B active
- **Architecture:** MoE · 192 experts, 6/token · DeepseekV41ForCausalLM
- **Context:** 1,048,576 tokens
- **Task:** image-text-to-text · vision
- **Signal:** 2 HF likes · 37 downloads

## Runs on

mixed: real files + estimate; sizing calculation, not tested.

- Q4 ~357.9 GB: Mac 512GB
- Q8 ~523.8 GB (estimated): no listed device
- FP16 ~984.5 GB (estimated): no listed device

## Links

- [Hugging Face](https://huggingface.co/GrootTeam/Groot-280B-A16B-261023)