Downloads · 30 days
591
25% of all-time downloads
just1moremodel/Qwen3.8-27B-Uncensored-MXFP4-awq
Qwen3.8-27B-Uncensored-MXFP4-awq is a text generation model from just1moremodel. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
MXFP4 quantization of orcarouter/Qwen3.8-27B-Uncensored, made with AMD Quark and exported as an AWQ-style checkpoint for serving with vLLM on AMD RDNA4 GPUs (gfx1201 / RX 9000 series) with ROCm.
Downloads · 30 days
591
25% of all-time downloads
All-time downloads
2.3K
Public
Parameters
15.6B
19.4 GB on disk
Likes
12
Public
Click a slice to open those files.
.safetensors19.4 GB · 100%
How the weights are stored.
U812.2B · 78%
From the Hugging Face model README
MXFP4 quantization of orcarouter/Qwen3.8-27B-Uncensored, made with AMD Quark and exported as an AWQ-style checkpoint for serving with vLLM on AMD RDNA4 GPUs (gfx1201 / RX 9000 series) with ROCm.
Start with this image: https://hub.docker.com/r/stilldeadcode/vllm-radiance
Add this if needed (needed at the time of testing): https://codeberg.org/ggz14/radiance-vllm-mxfp4
NOTE: fp8_mtp.py was already ran on this model, it is not needed to run again.
Ran Livebench Math for quality check (GLM 5.3 was grader): Final scoreboard (official 182q math, xhigh, 150k tokens) AMPS_Hard 77.0% math_comp 97.7% olympiad 100% Overall 86.5%
| Source model | orcarouter/Qwen3.8-27B-Uncensored |
| Architecture | Qwen3.5-27B (Qwen3_5ForConditionalGeneration) |
| Quantization tool | AMD Quark |
| Quantization | MXFP4 (Quark export, quant_method: quark) |
| KV cache | Post-RoPE KV quantization enabled |
| Context length | 262,144 |
| Hidden size / layers | 5120 / 64 |
| Attention | 24 heads, 4 KV heads (GQA) |
| Vocab size | 248,320 |
| Multimodal | Vision tower present (excluded from quantization) |
Quantization excludes lm_head and the vision tower, so those run at higher precision.
Serve with vLLM:
vllm serve just1moremodel/Qwen3.8-27B-Uncensored-MXFP4-awq \
--port 8081
Requires a vLLM build with Quark/MXFP4 support (ROCm on RDNA4 recommended).
model.safetensors (~19 GB), sized to fit consumer GPUs with 24 GB+ VRAM (depending on context length and offloading).Apache-2.0, inherited from the base model.