AR

AREX 2

ArexGGUF by mradermacherUpdated 2026-10-01
Download on Hugging Face
Parameters27B
Context256K
Layers64
Architectureqwen35
Smallest quant10.7 GB
Quick answer

AREX 2 needs about 19.2 GB of memory at Q4_K_M (a 16.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 10.7 GB (Q2_K) to 28.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 8.6 GB.

AREX 2 – 27B multimodal model with long‑horizon self‑improvement

AREX 2 is a 27‑billion‑parameter Qwen3.8‑compatible multimodal model from BAAI. It supports reasoning, tools, and vision, and is trained on algorithmic‑programming and machine‑learning tasks. The model can run up to 262,144 tokens of context.

Good at

  • Long‑horizon self‑improvement
  • Reasoning with feedback
  • Multimodal vision support

Limits

  • Requires large VRAM for full precision
  • No native vision support in llama.cpp
  • Apache‑2.0 license limits commercial use

Best for

  • Iterative coding refinement
  • Machine‑learning model debugging
  • Agentic reasoning tasks

Who it’s for. Developers and researchers needing a multimodal agent that can improve solutions over multiple rounds. The model is suitable for tasks that benefit from long‑horizon reasoning and feedback‑driven iteration.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K10.7 GB 13.4 GB
Too big Get file →
Q3_K_S12.1 GB 14.7 GB
Too big Get file →
Q3_K_M13.3 GB 15.9 GB
Too big Get file →
Q3_K_L14.3 GB 17.0 GB
Too big Get file →
IQ4_XS15.2 GB 17.8 GB
Too big Get file →
Q4_K_S15.6 GB 18.2 GB
Too big Get file →
Q4_K_M16.5 GB 19.2 GB
Too big Get file →
Q5_K_S18.7 GB 21.3 GB
Too big Get file →
Q5_K_M19.2 GB 21.9 GB
Too big Get file →
Q6_K22.1 GB 24.7 GB
Too big Get file →
Q8_028.6 GB 31.2 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_S (15.59 GB file, ~18.2 GB memory at 8 k context). For 16 GB VRAM, Q5_K_S (18.68 GB file, ~21.3 GB memory). For 24 GB VRAM, Q6_K (22.08 GB file, ~24.7 GB memory). Estimates are approximate.

What GPU do you need for AREX 2?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q3_K_M (13.3 GB file, about 15.9 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (19.2 GB file, about 21.9 GB in use).
  • 32 GB VRAM: the best quant that fits is Q8_0 (28.6 GB file, about 31.2 GB in use).

Which AREX 2 quant should you choose?

Q4_K_M is the usual starting point for AREX 2: it balances file size and quality well. Here it is a 16.5 GB file.

The Q8_0 file is 28.6 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 10.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters27B
Layers64
Trained context262,144 tokens
KV cache (fp16)256 KB per token

Source

Quantized bymradermacher
FamilyArex
Base modelBAAI/AREX-2
Licenseapache-2.0
Downloads (HF)8,796
Last updated2026-10-01

Frequently asked questions about AREX 2

How much VRAM does AREX 2 need?

About 19.2 GB at Q4_K_M with an 8K context window. That is the 16.5 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run AREX 2 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 13.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of AREX 2 GGUF?

It depends on the quant: Q2_K is 10.7 GB, Q3_K_S is 12.1 GB, Q3_K_M is 13.3 GB, Q3_K_L is 14.3 GB, IQ4_XS is 15.2 GB, Q4_K_S is 15.6 GB and 5 more.

What context length does AREX 2 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.1 GB at 8K and 8.6 GB at 32K.

Where can I download AREX 2 GGUF?

The files are on Hugging Face in the mradermacher/AREX-2-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does AREX 2 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run AREX 2 on 8 GB VRAM?

Use the Q4_K_S quant, which requires about 18.2 GB of VRAM at 8 k context. 8 GB is insufficient for full precision.

Does AREX 2 support vision?

Yes, the model has detected vision ability, but llama.cpp must be compiled with multimodal support.

What license applies to AREX 2?

The model is released under the Apache‑2.0 license, allowing commercial use with attribution.

Other GGUFs of AREX-2

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Arex models

All Arex →

Other medium models

All medium models →