Qwen3.8-27B-Uncensored-GGUF
Uncensored Qwen3.8-27B, published as GGUF quantizations with the multi-token prediction (MTP) head retained and verified.
Params
27.0 B
Context
—
Downloads 30d
145 K
Likes
3
Download history
tracking started — chart appears after 7 days of snapshots (1 recorded)145 K downloads in the last 30 days
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| Qwen3.8-27B-Uncensored-vision-f16.gguf | GGUF | 0.9 GB | 5.6 GB | ✅ Runs comfortably |
| Qwen3.8-27B-Uncensored-IQ2_M.gguf | IQ2_M | 10.6 GB | 16.2 GB | ✅ Runs comfortably |
| Qwen3.8-27B-Uncensored-IQ4_XS.gguf | IQ4_XS | 15.3 GB | 21.4 GB | ⚠️ Tight — reduce context |
| Qwen3.8-27B-Uncensored-Q4_K_M.gguf | Q4_K_M | 16.8 GB | 23.0 GB | ⚠️ Tight — reduce context |
| Qwen3.8-27B-Uncensored-Q5_K_M.gguf | Q5_K_M | 19.5 GB | 26.0 GB | ❌ Won’t fit |
| Qwen3.8-27B-Uncensored-Q6_K.gguf | Q6_K | 22.4 GB | 29.2 GB | ❌ Won’t fit |
| Qwen3.8-27B-Uncensored-Q8_0.gguf | Q8_0 | 29.0 GB | 36.5 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run qwen3-8-27b-uncensored-gguf-azukivc # pin the quantization explicitly $ ollama run qwen3-8-27b-uncensored-gguf-azukivc-q4_k_m
$ huggingface-cli download azukivc/Qwen3.8-27B-Uncensored-GGUF-GGUF \
Qwen3.8-27B-Uncensored-Q4_K_M.gguf --local-dir .
$ llama-cli -m Qwen3.8-27B-Uncensored-Q4_K_M.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/qwen3-8-27b-uncensored-gguf-azukivc
{
"hf_id": "azukivc/Qwen3.8-27B-Uncensored-GGUF",
"params_b": 27.00,
"context_length": null,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 145414,
"vram_estimates": [
{ "quant": "IQ2_M", "gb": 16.2 },
{ "quant": "IQ4_XS", "gb": 21.4 }
],
"updated_at": "2026-10-25T01:00:27Z"
}
Specifications
- Parameters
- 27.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-08-22
- Base model
- Qwen3.8-27B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-25
Compare with any text-generation model