Nex N2.5 Mini Un needs about 17.7 GB of memory at its main file (a 16.5 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 4 quants, from 13.5 GB to 25.3 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
Nex N2.5 Mini Uncensored APEX GGUF – 34.7B MoE model with native vision support
Nex N2.5 Mini Uncensored APEX GGUF is a 34.7B parameter mixture‑of‑experts model based on qwen35moe architecture. It supports up to 262,144‑token context and native vision. The model is released under Apache‑2.0 and is quantized with APEX for MoE awareness.
Good at
- Large‑context generation
- Native vision processing
- Mixture‑of‑experts efficiency
Limits
- No vision support in non‑APEX frontends
- Requires recent llama.cpp
- Large VRAM needed for full context
Best for
- Long‑form content creation
- Multimodal inference
- Research prototyping
Who it’s for. Developers and researchers needing a high‑capacity MoE model with vision capabilities. The model is suitable for applications that can accommodate its memory footprint and require the flexibility of expert routing.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 13.5 GB | 14.6 GB | Too big | Get file → | |
| — | 16.5 GB | 17.7 GB | Too big | Get file → | |
| — | 22.8 GB | 24.0 GB | Too big | Get file → | |
| — | 25.3 GB | 26.5 GB | Too big | Get file → |
For a 16‑GB VRAM budget, use the 13.47‑GB file (est. 14.6‑GB memory at 8k context). For 32‑GB VRAM, the 16.54‑GB file (est. 17.7‑GB memory) is suitable. Larger VRAM allows the 22.82‑GB or 25.34‑GB files (est. 24‑26.5‑GB memory).
What GPU do you need for Nex N2.5 Mini Un?
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 the main file (13.5 GB file, about 14.6 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (22.8 GB file, about 24.0 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (25.3 GB file, about 26.5 GB in use).
Which Nex N2.5 Mini Un quant should you choose?
If memory is tight, the smallest file is the first one in the table at 13.5 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
Source
Frequently asked questions about Nex N2.5 Mini Un
How much VRAM does Nex N2.5 Mini Un need?
About 17.7 GB at its main file 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 Nex N2.5 Mini Un on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 14.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Nex N2.5 Mini Un GGUF?
It depends on the quant: file is 13.5 GB, file is 16.5 GB, file is 22.8 GB, file is 25.3 GB.
What context length does Nex N2.5 Mini Un support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download Nex N2.5 Mini Un GGUF?
The files are on Hugging Face in the vikrant1123/Nex-N2.5-mini-Uncensored-APEX-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 Mini Un 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 this model on 8 GB VRAM?
The smallest file requires about 14.6 GB of VRAM at 8k context, so 8 GB is insufficient.
Does this model support vision?
Yes, it has native vision support as part of its architecture.
What license applies to this model?
The model is released under the Apache‑2.0 license.