NE

Nex N2.5 Pro

NexGGUF by bartowskiUpdated 2026-09-15
Download on Hugging Face
Parameters396B
Context256K
Layers60
Architectureqwen35moe
Smallest quant1.1 GB
Quick answer

Nex N2.5 Pro needs about 252.3 GB of memory at Q4_K_M (a 250.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 25 quants, from 1.1 GB to 421.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.0 GB.

Nex N2.5 Pro – a 397B‑parameter, vision‑enabled agentic model

Nex N2.5 Pro is a 397B‑parameter model built on the qwen35moe architecture. It supports vision and long‑horizon tasks in real‑world environments. The model is released under the Apache‑2.0 license and is available in multiple quantizations for llama.cpp.

Good at

  • Vision‑guided agentic tasks
  • Long‑horizon planning
  • Computer and web browsing

Limits

  • No speculative decoding
  • Requires large VRAM for high‑precision quants
  • No text‑only support for vision tasks

Best for

  • Autonomous program execution
  • Scientific research workflows
  • Complex productivity tasks

Who it’s for. Developers and enterprises looking to deploy agentic models with visual perception. The model is open source and can be integrated into custom applications or hosted services.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.1 GB 2.6 GB
Fits Get file →
IQ1_S87.9 GB (3 parts) 89.4 GB
Too big Get file →
IQ1_M98.1 GB (3 parts) 99.6 GB
Too big Get file →
IQ2_XXS115.1 GB (3 parts) 116.6 GB
Too big Get file →
IQ2_XS121.4 GB (4 parts) 122.9 GB
Too big Get file →
IQ2_S127.8 GB (4 parts) 129.3 GB
Too big Get file →
IQ2_M142.3 GB (4 parts) 143.8 GB
Too big Get file →
Q2_K150.8 GB (4 parts) 152.3 GB
Too big Get file →
IQ3_XXS169.5 GB (5 parts) 171.1 GB
Too big Get file →
IQ3_XS178.1 GB (5 parts) 179.6 GB
Too big Get file →
Q3_K_S180.0 GB (5 parts) 181.5 GB
Too big Get file →
Q3_K_M190.0 GB (5 parts) 191.5 GB
Too big Get file →
Q3_K_L200.2 GB (6 parts) 201.7 GB
Too big Get file →
IQ3_M209.4 GB (6 parts) 211.0 GB
Too big Get file →
IQ4_XS219.5 GB (6 parts) 221.0 GB
Too big Get file →
Q4_0225.4 GB (6 parts) 226.9 GB
Too big Get file →
Q4_K_S233.5 GB (6 parts) 235.0 GB
Too big Get file →
Q4_1249.1 GB (7 parts) 250.6 GB
Too big Get file →
IQ4_NL249.5 GB (7 parts) 251.0 GB
Too big Get file →
Q4_K_M250.8 GB (7 parts) 252.3 GB
Too big Get file →
Q5_K_S282.8 GB (8 parts) 284.3 GB
Too big Get file →
Q5_K_M305.6 GB (8 parts) 307.1 GB
Too big Get file →
Q6_K_S333.2 GB (9 parts) 334.7 GB
Too big Get file →
Q6_K349.8 GB (9 parts) 351.3 GB
Too big Get file →
Q8_0421.6 GB (11 parts) 423.1 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (250.83 GB file, 252.3 GB memory at 8 k context). For 16 GB VRAM, Q4_K_S (233.49 GB file, 235 GB memory) is a good balance. For 24 GB VRAM, Q3_K_M (189.95 GB file, 191.5 GB memory) offers higher precision with manageable memory usage. Estimates are approximate.

What GPU do you need for Nex N2.5 Pro?

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: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (1.1 GB file, about 2.6 GB in use).

Which Nex N2.5 Pro quant should you choose?

Q4_K_M is the usual starting point for Nex N2.5 Pro: it balances file size and quality well. Here it is a 250.8 GB file.

The Q8_0 file is 421.6 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 the first one in the table at 1.1 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters396B (approximate, mixture of experts)
Layers60
Trained context262,144 tokens
KV cache (fp16)120 KB per token

Source

Quantized bybartowski
FamilyNex
Licenseapache-2.0
Downloads (HF)2,862
Last updated2026-09-15

Frequently asked questions about Nex N2.5 Pro

How much VRAM does Nex N2.5 Pro need?

About 252.3 GB at Q4_K_M with an 8K context window. That is the 250.8 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 Nex N2.5 Pro on an 8 GB GPU?

Yes, with the smaller files: the 1.1 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Nex N2.5 Pro GGUF?

It depends on the quant: file is 1.1 GB, IQ1_S is 87.9 GB, IQ1_M is 98.1 GB, IQ2_XXS is 115.1 GB, IQ2_XS is 121.4 GB, IQ2_S is 127.8 GB and 19 more.

What context length does Nex N2.5 Pro support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.0 GB at 8K and 4.0 GB at 32K.

Where can I download Nex N2.5 Pro GGUF?

The files are on Hugging Face in the bartowski/Nex-N2.5-Pro-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 Nex N2.5 Pro 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 Nex N2.5 Pro on 8 GB VRAM?

Only the lowest‑precision quants such as Q4_K_M or Q4_K_S can fit, and even then memory usage will be close to the VRAM limit. Higher‑precision quants require more than 8 GB.

Does the model support vision?

Yes, Nex N2.5 Pro has detected vision capability and can process visual inputs.

What license does the model use?

The model is released under the Apache‑2.0 license.

Other GGUFs of Nex-N2.5-Pro

Similar models

Browse all →

More from bartowski

All by bartowski →

More Nex models

All Nex →

Other large models

All large models →