Qwen3.8 Flash Next UNCENSORED IQ2 XXS needs about 76.0 GB of memory at IQ2_XXS (a 74.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.
Qwen3.8 Flash Next UNCENSORED IQ2 XXS – 177B parameter model quantized for 64 GB RAM and 16 GB GPU
This model is a 177 billion‑parameter Qwen4exp architecture with 262,144 trained context tokens. It is a quantized IQ2_XXS GGUF conversion of the UNCENSORED FP8 checkpoint, designed for systems with 64 GB RAM and a 16 GB GPU. The model removes safety refusals and improves reasoning depth compared to the base FP8 version.
Good at
- Reasoning depth improvement
- Zero hard refusals on HarmBench
- High token throughput with IQ2_XXS quantization
Limits
- No vision support
- Requires recent llama.cpp with qwen4exp support
- License governed by Qwen Community License 1.0
Best for
- Large‑context reasoning tasks
- Applications needing 262,144 token context
- Deployments on 64 GB RAM systems
Who it’s for. Engineers and researchers building LLM applications that require extensive context and reasoning capabilities without safety refusal behavior. Suitable for developers with access to a 16 GB GPU and 64 GB RAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 74.7 GB | 76.0 GB | Too big | Get file → |
For 64 GB RAM use the IQ2_XXS quantized file (74.66 GB). The estimated VRAM requirement at 8 k context is 76 GB, so a 16 GB GPU will need off‑load or CPU‑only inference. If GPU memory is limited, consider a higher‑bit quant like IQ4_K_M, though file size and memory will increase.
What GPU do you need for Qwen3.8 Flash Next UNCENSORED IQ2 XXS?
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: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Qwen3.8 Flash Next UNCENSORED IQ2 XXS quant should you choose?
IQ2_XXS (74.7 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Qwen3.8 Flash Next UNCENSORED IQ2 XXS
How much VRAM does Qwen3.8 Flash Next UNCENSORED IQ2 XXS need?
About 76.0 GB at IQ2_XXS with an 8K context window. That is the 74.7 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 UNCENSORED IQ2 XXS on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 76.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Qwen3.8 Flash Next UNCENSORED IQ2 XXS GGUF?
It depends on the quant: IQ2_XXS is 74.7 GB.
What context length does Qwen3.8 Flash Next UNCENSORED IQ2 XXS support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.2 GB at 32K.
Where can I download Qwen3.8 Flash Next UNCENSORED IQ2 XXS GGUF?
The files are on Hugging Face in the BattleGhost/Qwen3.8-Flash-Next-UNCENSORED-IQ2_XXS-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 UNCENSORED IQ2 XXS use?
Hugging Face lists the license as other. 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?
No, the model requires 76 GB VRAM at 8 k context. A 16 GB GPU will need CPU off‑load or a lower‑bit quant.
What context length does it support?
It supports up to 262,144 tokens, the trained context length of the base model.
Is the model safe to use?
The safety refusals have been removed; users must ensure compliance with laws and platform policies.