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

DexopT/Qwen2.5-Coder-3B-Instruct-abliterated-MNN

Name: DexopT/Qwen2.5-Coder-3B-Instruct-abliterated-MNN
Author: DexopT

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/DexopT%2FQwen2.5-Coder-3B-Instruct-abliterated-MNN"

Response includes

classification m1
files 9
hub_downloads_all_time 364
author_summary 4 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

364

105 last 30d - stable

Likes

Model age

8mo ago

created 2026-01-27

Downloads over time

Mar 18 → Oct 11 · 69 snapshots · spans 207 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.

WhoIsShe/Qwen2.5-Coder-3B-Instruct-abliterated-MNN this lineage

DexopT/Qwen2.5-Coder-3B-Instruct-abliterated-MNN↓ 105

Metadata

License

other

Languages

Files by quantization

Auxiliary files 9 files 1.80 GB

llm.mnn.weight	1.80 GB	08a73cda	download
tokenizer.txt	3.05 MB	c77936bb	download
llm.mnn.json	1.20 MB	0d3aab23	download
llm.mnn	613 KB	5e524280	download
llm_config.json	2.86 KB	2528331d	download
.gitattributes	1.58 KB	e547efe9	download
README.md	1.19 KB	fdf25ee0	download
export_args.json	1.07 KB	633128c3	download
config.json	210 B	bd685f44	download

README current version from Hugging Face

language:

en
license: other
tags:
mnn
on-device
android
ios
quantization
int4
text-generation
qwen-2.5
pipeline_tag: text-generation
library_name: mnn
base_model: WhoIsShe/Qwen2.5-Coder-3B-Instruct-abliterated-MNN

huihui-ai/Qwen2.5-Coder-3B-Instruct-abliterated (MNN Quantized)

Original model :

https://huggingface.co/huihui-ai/Qwen2.5-Coder-3B-Instruct-abliterated

This is a 4-bit quantized version of the huihui-ai/Qwen2.5-Coder-3B-Instruct-abliterated, optimized for on-device inference (Android/iOS) using the Alibaba MNN framework.

🚀 Fast Deployment on Android

1. Download the App

Don't build from scratch! Use the official MNN Chat Android app:

Download APK (GitHub)

2. Setup

Download the files from this repo (llm.mnn, llm.mnn.weight, config.json).
Create a folder on your phone: /sdcard/MNN/Qwen2.5-Coder-3B-Instruct-abliterated.
Copy the files into that folder.
Open the MNN App and select your folder.

💻 Technical Details

Framework: MNN
Quantization: 4-bit Asymmetric (Int4)
Model Type: QWEN-2.5-CODER (Uncensored)

README history 1 version

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

2026-01-27Add MNN 4-bit quantized model with Model Carded203181.2 KB
Loading...