← back to catalog · registered 2026-08-22 13:56

MegaPanchamZ/Qwen3.8-9B-abliterated-25-GGUF

Name: MegaPanchamZ/Qwen3.8-9B-abliterated-25-GGUF
Author: MegaPanchamZ

Your rig guess connected

? Why do I need an app?

Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/MegaPanchamZ%2FQwen3.8-9B-abliterated-25-GGUF"

Response includes

classification m8
files 3
hub_downloads_all_time 7,709
author_summary 2 models
readme_text full

10 credits · hourly refresh · ~4 KB payload Get an API key →

Abliteration classifier · v1.0.0

Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.

Confidence

MEDIUM

Inherited from base model

M1 Direct removal →

Why this label 3 signals

Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.

'abliterated' in name/tags
is_gguf=1
assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown

Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

Downloads · lifetime

5K last 30d - active

Likes

Model age

7w ago

created 2026-08-19

Downloads over time

Aug 19 → Oct 11 · 48 snapshots · spans 53 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

MegaPanchamZ/Qwen3.8-9B-abliterated-25↓ 352 this lineage

MegaPanchamZ/Qwen3.8-9B-abliterated-25-GGUF↓ 5,395

Variants by this author 2 formats · 6K downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

MegaPanchamZ/Qwen3.8-9B-abliterated-25-GGUF (this) GGUF 5K
MegaPanchamZ/Qwen3.8-9B-abliterated-25 safetensors 352

Metadata

License

apache-2.0

Quantizations

Q4_K

Files by quantization

Q4_K 1 file 5.24 GB

Qwen3.8-9B-abliterated-25.Q4_K_M.gguf

5.24 GB

542d41f6

download

Auxiliary files 2 files 4.42 KB

README.md	2.87 KB	027efa2a	download
.gitattributes	1.56 KB	ed66b403	download

README current version from Hugging Face

license: apache-2.0
base_model: MegaPanchamZ/Qwen3.8-9B-abliterated-25
tags:

gguf
llama.cpp
abliteration
heretic
quantized
qwen3.5
distillation
reasoning
text-generation
en
model_creator: MegaPanchamZ
quantized_by: MegaPanchamZ

Qwen3.8-9B Abliterated (Q4_K_M)

A GGUF Q4_K_M quantization of MegaPanchamZ/Qwen3.8-9B-abliterated-25, the abliterated (decensored) version of empero-ai/Qwen3.8-9B.

Abliteration was performed with Heretic v1.4.0 (GPLv3+, by Philipp Emanuel Weidmann), which ablates refusal directions from attn.o_proj and mlp.down_proj via LoRA adapters, then merges them back into the base weights.

Results

Metric	Original	Abliterated
Refusals (100 harmful prompts)	99/100	25/100
KL divergence vs. original	—	0.0142

25/100 refusals with a very low KL divergence (0.014, far below the 0.5 damage threshold) — strong refusal suppression with minimal impact on model capabilities.

Abliteration details

Tool: Heretic v1.4.0 (auto batch size 64, 400 total optimization trials)
Prompt sets: mlabonne/harmless_alpaca (good) and mlabonne/harmful_behaviors (bad), 400 prompts each
Selected trial: 276 of 400 (Pareto-optimal)
Parameters:
- direction_index = 17.52
- attn.o_proj.max_weight = 1.39
- attn.o_proj.max_weight_position = 19.72
- attn.o_proj.min_weight = 1.36
- attn.o_proj.min_weight_distance = 9.88
- mlp.down_proj.max_weight = 1.27
- mlp.down_proj.max_weight_position = 26.42
- mlp.down_proj.min_weight = 1.25
- mlp.down_proj.min_weight_distance = 17.97
Method: LoRA-based row-normalized ("full") ablation, merged into base weights (no separate adapters)
Format: Merged safetensors → GGUF Q4_K_M via Unsloth save_pretrained_gguf

Quantization

Q4_K_M (5.3 GB), converted with Unsloth's llama.cpp toolchain
Quantization applied to the merged, abliterated bf16 weights

Usage (llama.cpp)

llama-server -m Qwen3.8-9B-abliterated-25.Q4_K_M.gguf -ngl 99

[!NOTE]
This is a reasoning model — responses start with a <think> block. Use generous max_tokens (1000+) when chatting, or the answer may be truncated.

License and attribution

Base model: Apache-2.0 (empero-ai/Qwen3.8-9B)
Abliterated with Heretic v1.4.0 — https://heretic-project.org — GPLv3+ (source)
GGUF conversion: Unsloth

Disclaimer

This model has had its refusal behavior modified. It may comply with requests the original model would decline. Use responsibly and in accordance with local laws.

README history 2 versions

The author's README evolved over time. Click a version to see its content at that point.

2026-08-19Upload README.md with huggingface_hubd9d405a2.9 KB
Loading...
2026-08-19Upload README.md with huggingface_hub3a8f5a32.7 KB
Loading...