---
title: BAAI/AREX-2
url: https://recentmodel.com/models/BAAI/AREX-2/
status: archived
published: 2026-09-29T17:56:21Z
updated: 2026-10-25T11:20:32Z
source: https://huggingface.co/BAAI/AREX-2
---

# BAAI/AREX-2

AREX-2 is a 27B-parameter long-horizon agent model from the Beijing Academy of Artificial Intelligence (BAAI). Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.


- **Org:** BAAI
- **Released:** 2026-09-29
- **Status:** archived
- **License:** apache-2.0 (permissive)
- **Parameters:** 27.4B
- **Architecture:** dense · Qwen3_5ForConditionalGeneration
- **Context:** 262,144 tokens
- **Task:** image-text-to-text · vision
- **Signal:** 128 HF likes · 2423 downloads
- **Archived:** 2026-10-25 (no longer listed by the collector; stats as last seen)

## Runs on

real quant files; sizing calculation, not tested.

- Q4 ~20.3 GB: 24GB GPU (3090 / 4090), 32GB GPU (5090), 24GB GPU (7900 XTX), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~31.8 GB: 48GB GPU (RTX 6000 Ada / A6000), Mac 48GB, Mac 64GB (M1 Max), CPU only, 64GB RAM (slow)
- FP16 ~56.9 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/BAAI/AREX-2)
- [GGUF: bartowski/BAAI_AREX-2-GGUF](https://huggingface.co/bartowski/BAAI_AREX-2-GGUF)
- [MLX: mlx-community/AREX-2-4bit](https://huggingface.co/mlx-community/AREX-2-4bit)
- [MLX: mlx-community/AREX-2-5bit](https://huggingface.co/mlx-community/AREX-2-5bit)
- [MLX: mlx-community/AREX-2-6bit](https://huggingface.co/mlx-community/AREX-2-6bit)
- [MLX: mlx-community/AREX-2-8bit](https://huggingface.co/mlx-community/AREX-2-8bit)