Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF
RVN is a double-refined abliterated variant of Qwen3.8-27B, built on top of trohrbaugh/Qwen3.8-27B-heretic-ara (an ARA abliteration by Tim Rohrbaugh) and further refined with two additional full-weight ARA passes targeting residual refusals. It retains very low behavioral damage (KL ≈ 0.0085) while reducing harmful-prompt refusals from 3/100 (source) to 0–1/100 in independent measurements.
Params
27.0 B
Context
—
Downloads 30d
1.1 M
Likes
867
Download history
daily snapshots · 83 days
▲ 85 K in the last 30 days (7.0%)
1.2 M1.0 M
Oct 12Oct 22Nov 1Nov 10
2.0 M245 K
Aug 20Sep 16Oct 14Nov 10
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 |
|---|---|---|---|---|
| RVN-IQ1_S-multilingual-mtp.gguf | IQ1_S | 7.6 GB | 12.9 GB | ✅ Runs comfortably |
| RVN-IQ1_M-multilingual-mtp.gguf | IQ1_M | 8.1 GB | 13.4 GB | ✅ Runs comfortably |
| RVN-IQ2_XXS-multilingual-mtp.gguf | IQ2_XXS | 8.9 GB | 14.3 GB | ✅ Runs comfortably |
| RVN-IQ2_XS-multilingual-mtp.gguf | IQ2_XS | 9.5 GB | 15.0 GB | ✅ Runs comfortably |
| RVN-IQ2_S-multilingual-mtp.gguf | IQ2_S | 9.8 GB | 15.3 GB | ✅ Runs comfortably |
| RVN-IQ2_M-multilingual-mtp.gguf | IQ2_M | 10.5 GB | 16.1 GB | ✅ Runs comfortably |
| RVN-Q4_K_M-multilingual-vision.gguf | Q4_K_M | 18.7 GB | 25.2 GB | ❌ Won’t fit |
| model.safetensors | fp16 | 28.4 GB | 35.8 GB | ❌ Won’t fit |
| RVN-BF16-mtp.gguf | GGUF | 54.3 GB | 64.2 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-heretic-abliterated-uncensored-gguf # pin the quantization explicitly $ ollama run qwen3-8-27b-heretic-abliterated-uncensored-gguf-q4_k_m
$ huggingface-cli download 0bserverx/Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF-GGUF \
RVN-Q4_K_M-multilingual-vision.gguf --local-dir .
$ llama-cli -m RVN-Q4_K_M-multilingual-vision.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/qwen3-8-27b-heretic-abliterated-uncensored-gguf
{
"hf_id": "0bserverx/Qwen3.8-27B-Heretic-Abliterated-Uncensored-GGUF",
"params_b": 27.00,
"context_length": null,
"license": { "id": "apache-2.0", "commercial": "yes" },
"downloads_30d": 1124501,
"vram_estimates": [
{ "quant": "Q4_K_M", "gb": 25.2 },
{ "quant": "GGUF", "gb": 64.2 }
],
"updated_at": "2026-09-21T01:00:04Z"
}
Specifications
- Parameters
- 27.0 B
- Licence
- apache-2.0
- First seen on the Hub
- 2026-08-14
- Base model
- Qwen3.8-27B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-08-20
Compare with any text-generation model