Nex N2.5 mini OrcaRouter Sangreal ICE needs about 25.7 GB of memory at its main file (a 22.5 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 7 quants, from 11.9 GB to 69.4 GB (BF16).
35B Qwen3.5-MoE model with OrcaRouter trunk and Sangreal ICE quantization
Nex N2.5 mini OrcaRouter Sangreal ICE is a 35‑billion parameter Qwen3.5‑MoE model. It uses an abliterated OrcaRouter trunk and is quantized with Sangreal ICE. The model supports a 262,144‑token context and is released under Apache‑2.0.
Good at
- Large context handling
- Mixture‑of‑Experts routing
- Uncensored agentic behavior
Limits
- No vision support in base GGUF
- Requires high VRAM for full‑precision
- Uncensored – sandboxing recommended
Best for
- Long‑form content generation
- Agentic task orchestration
- Research on MoE architectures
Who it’s for. Developers and researchers needing a large, uncensored Qwen3.5‑MoE model with extended context. Suitable for experimentation with OrcaRouter routing and Sangreal calibration.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 11.9 GB | 13.8 GB | Too big | Get file → | |
| — | 14.5 GB | 16.8 GB | Too big | Get file → | |
| — | 18.5 GB | 21.2 GB | Too big | Get file → | |
| — | 22.5 GB | 25.7 GB | Too big | Get file → | |
| — | 24.5 GB | 28.0 GB | Too big | Get file → | |
| Q8_0 | 36.9 GB | 41.8 GB | Too big | Get file → | |
| BF16 | 69.4 GB (2 parts) | 78.2 GB | Too big | Get file → |
For 8 GB VRAM use the 12G‑ICEv2 tier (file size 11.9 GB, ~13.8 GB memory). For 16 GB VRAM the 15G‑ICEv2 tier (14.51 GB, ~16.8 GB). For 32 GB VRAM the 25G‑ICEv2 tier (24.53 GB, ~28 GB).
What GPU do you need for Nex N2.5 mini OrcaRouter Sangreal ICE?
These figures add the file size, an fp16 KV cache at an 8K context window 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 (11.9 GB file, about 13.8 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (18.5 GB file, about 21.2 GB in use).
- 32 GB VRAM: the best quant that fits is the main file (24.5 GB file, about 28.0 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (36.9 GB file, about 41.8 GB in use).
Which Nex N2.5 mini OrcaRouter Sangreal ICE quant should you choose?
If memory is tight, the smallest file is the first one in the table at 11.9 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 OrcaRouter Sangreal ICE
How much VRAM does Nex N2.5 mini OrcaRouter Sangreal ICE need?
About 25.7 GB at its main file with an 8K context window. That is the 22.5 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 OrcaRouter Sangreal ICE on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 13.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Nex N2.5 mini OrcaRouter Sangreal ICE GGUF?
It depends on the quant: file is 11.9 GB, file is 14.5 GB, file is 18.5 GB, file is 22.5 GB, file is 24.5 GB, Q8_0 is 36.9 GB and 1 more.
What context length does Nex N2.5 mini OrcaRouter Sangreal ICE support?
The GGUF header lists a trained context length of 262,144 tokens (256K).
Where can I download Nex N2.5 mini OrcaRouter Sangreal ICE GGUF?
The files are on Hugging Face in the gbuzhf/Nex-N2.5-mini-OrcaRouter-Sangreal-ICE 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 OrcaRouter Sangreal ICE 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?
Use the 12G‑ICEv2 tier; it requires about 13.8 GB of VRAM at 8K context, which is close to 8 GB but may need swap or lower context.
Does this model support vision?
Vision is available only through the separate projector file mmproj‑Nex‑N2.5‑mini‑OrcaRouter‑Q8_0.gguf, not in the base GGUF.
Is the model censored?
No, the model is uncensored; it has no refusal backstop and should be sandboxed at the OS level.