Qwen3.8 Flash Next GSQ RCO abliterated needs about 75.3 GB of memory at Q2_0 (a 66.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 0.9 GB (Q2_0) to 83.9 GB (IQ3_S).
Qwen3.8-Flash-Next GSQ RCO abliterated – a 177B parameter sparse Mixture‑of‑Experts model
This GGUF model is a 177 billion‑parameter Qwen4exp‑mtp architecture with 512 routed experts per layer. It uses non‑uniform GSQ‑RCO quantization and includes a vision projector for multimodal use. The model is released under the Apache‑2.0 license.
Good at
- Sparse expert inference
- Multimodal vision tasks
- Large‑scale language modeling
Limits
- No built‑in guardrails – refusal removed
- Requires 8 k context for memory estimates
- No vision support in quantized variants
Best for
- High‑throughput text generation
- Vision‑enabled inference with mmproj
- Research on sparse Mixture‑of‑Experts
Who it’s for. Engineers and researchers who need a large, sparse LLM with flexible quantization. The model is suitable for deployment in environments that can handle the memory requirements and lack of safety guardrails.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_0 | 0.9 GB | 1.5 GB | Fits | Get file → | |
| Q8_0 | 4.1 GB | 5.1 GB | Fits | Get file → | |
| Q2_0 | 66.8 GB (2 parts) | 75.3 GB | Too big | Get file → | |
| IQ2_XS | 68.5 GB (2 parts) | 77.2 GB | Too big | Get file → | |
| IQ3_XXS | 76.1 GB (2 parts) | 85.8 GB | Too big | Get file → | |
| IQ3_S | 83.9 GB (2 parts) | 94.4 GB | Too big | Get file → |
For 8 GB VRAM, use the Q2_0 quantization (file size 0.89 GB, estimated 1.5 GB memory at 8 k context). If more VRAM is available, Q8_0 (4.14 GB, 5.1 GB memory) provides higher fidelity. Larger budgets can use the 66.82 GB Q2_0 shard or the 68.47 GB IQ2_XS shard for full‑size performance.
What GPU do you need for Qwen3.8 Flash Next GSQ RCO abliterated?
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: the best quant that fits is Q2_0 (0.9 GB file, about 1.5 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 16 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 32 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
- 48 GB VRAM: the best quant that fits is Q8_0 (4.1 GB file, about 5.1 GB in use).
Which Qwen3.8 Flash Next GSQ RCO abliterated quant should you choose?
Q2_0 (66.8 GB) is the middle option in this repository. 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.
If memory is tight, the smallest file is Q2_0 at 0.9 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
Frequently asked questions about Qwen3.8 Flash Next GSQ RCO abliterated
How much VRAM does Qwen3.8 Flash Next GSQ RCO abliterated need?
About 75.3 GB at Q2_0 with an 8K context window. That is the 66.8 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 Qwen3.8 Flash Next GSQ RCO abliterated on an 8 GB GPU?
Yes, with Q8_0: the 4.1 GB file needs about 5.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Qwen3.8 Flash Next GSQ RCO abliterated GGUF?
It depends on the quant: Q2_0 is 0.9 GB, Q8_0 is 4.1 GB, Q2_0 is 66.8 GB, IQ2_XS is 68.5 GB, IQ3_XXS is 76.1 GB, IQ3_S is 83.9 GB.
Where can I download Qwen3.8 Flash Next GSQ RCO abliterated GGUF?
The files are on Hugging Face in the KilltheSniper/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-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 Qwen3.8 Flash Next GSQ RCO abliterated 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?
Yes, the Q2_0 quantization uses about 1.5 GB of VRAM at 8 k context, which fits within 8 GB. The Q8_0 quantization uses about 5.1 GB and also fits.
Does this model support vision tasks?
Yes, the model includes a vision projector (mmproj) at BF16 for multimodal use.
What license applies to this model?
The model is released under the Apache‑2.0 license.