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

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

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

Response includes

classification m1
files 13
hub_downloads_all_time 668
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

Direct removal

No other method signals detected in this model.

Confidence

MEDIUM

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=0 (base model)
no specific method indicators - defaulting to M1 (most common)

Refusal direction extracted via

Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗

Confidence

MEDIUM

Why we say so

primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)

Downloads · lifetime

668

352 last 30d - active

Likes

Descendants

in 3 direct forks

Model age

7w ago

created 2026-08-19

Downloads over time

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

Genealogy 3 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.

empero-ai/Qwen3.8-9B this lineage

MegaPanchamZ/Qwen3.8-9B-abliterated-25↓ 352 3 forks

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

nuofang/Qwen3.8-9B-abliterated-25-GGUF↓ 2,377

Krypto-Whitehat/qwen3.8-9b-uncensored-cyber-xrpl-v4↓ 250

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

Metadata

License

apache-2.0

Files by quantization

Auxiliary files 13 files 17.5 GB

model-00002-of-00004.safetensors	4.65 GB	1cc84ae7	download
model-00003-of-00004.safetensors	4.61 GB	ffa06701	download
model-00001-of-00004.safetensors	4.60 GB	576aa9ec	download
model-00004-of-00004.safetensors	3.66 GB	99567908	download
tokenizer.json	19.1 MB	06b95093	download
model.safetensors.index.json	67.6 KB	778b7bbb	download
chat_template.jinja	7.57 KB	a585dec8	download
tokenizer_config.json	7.02 KB	ecc67d85	download
config.json	2.82 KB	6eb64936	download
README.md	2.79 KB	27c122d7	download
.gitattributes	1.53 KB	52373fe2	download
processor_config.json	1.16 KB	33818c7f	download
generation_config.json	164 B	0d348acb	download

README current version from Hugging Face

license: apache-2.0
base_model: empero-ai/Qwen3.8-9B
tags:

safetensors
transformers
abliteration
heretic
qwen3.5
distillation
reasoning
text-generation
en
model_creator: MegaPanchamZ

Qwen3.8-9B Abliterated

An abliterated (decensored) version of empero-ai/Qwen3.8-9B, exported as merged safetensors.

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.

A GGUF Q4_K_M quantization is available at MegaPanchamZ/Qwen3.8-9B-abliterated-25-GGUF.

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, bf16, 4 shards (~5 GB each)

Usage (transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("MegaPanchamZ/Qwen3.8-9B-abliterated-25", torch_dtype="auto")
tokenizer = AutoTokenizer.from_pretrained("MegaPanchamZ/Qwen3.8-9B-abliterated-25")

[!NOTE]
This is a reasoning model — responses start with a thinking 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)

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 1 version

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

2026-08-19Upload abliterated model (Heretic v1.4.0)9b903272.8 KB
Loading...