huihui-ai / Model details

Huihui-Qwen3-VL-8B-Instruct-abliterated

huihui-ai/Huihui-Qwen3-VL-8B-Instruct-abliterated

abliterateduncensoredfine-tuned

Read the original model card ↗ · Choosing a model

About this model

Huihui's Qwen3-VL-8B-Instruct derivative targets refusals in image-and-text conversations through abliteration. The creator says only the language component was modified; the image component was left untouched. Its stated aim is to reduce refusals to describe or analyze images, not to retrain visual recognition.

Description sources

https://huggingface.co/huihui-ai/Huihui-Qwen3-VL-8B-Instruct-abliterated ↗

Source-based description, not an independent evaluation. Check the linked sources for the creator’s claims and limitations.

Size and formats

Base model: Qwen/Qwen3-VL-8B-Instruct
Parameters: 8.76712B
Main repository formats: safetensors
License tag: apache-2.0
Languages: Unknown
Last updated: 2025-12-15 · 07:43:55.000 UTC

What changed

Model types are creator-reported, not independently tested. These labels describe source metadata, not a guarantee of behavior.

abliteratedsource tag: abliterated
abliteratedmodel name: Huihui-Qwen3-VL-8B-Instruct-abliterated
uncensoredsource tag: uncensored
fine-tunedsource tag: base_model:finetune:Qwen/Qwen3-VL-8B-Instruct

Family evidence

These groupings are source-reported. Explicit base identifiers take priority over generic architecture tags.

Qwen: Base model: Qwen/Qwen3-VL-8B-Instruct

Available versions

Versions appear here only when they explicitly name this exact model as their source. Check each repository for available files.

Refusal profile

Refused 1 of 120 prompts (0.8%)

We asked this model 120 benign but risky-looking prompts across ten topics and counted how often it refused, hedged or answered. Lower means it answered more often; it says nothing about answer quality.

Persuasion & deception0 refused · 0 partial · 12 complied
Harassment & confrontation0 refused · 0 partial · 12 complied
Risky activities0 refused · 0 partial · 12 complied
Offensive & controversial speech0 refused · 0 partial · 12 complied
Legal grey areas0 refused · 0 partial · 12 complied
Privacy & personal data0 refused · 0 partial · 12 complied
Self-harm & mental health topics0 refused · 0 partial · 12 complied
Sexual & adult topics1 refused · 0 partial · 11 complied
Ethically grey requests0 refused · 0 partial · 12 complied
Violence & conflict0 refused · 0 partial · 12 complied

Capability check: 14 of 15 passed. Fifteen simple tasks with known answers, so a broken model cannot score well by refusing nothing.

Unclear answers: 0 (not counted). An answer is unclear when the two checker models disagree or the output was empty or garbled.

Measured 2026-10-24 · Q4_K_M via mradermacher/Huihui-Qwen3-VL-8B-Instruct-abliterated-GGUF ↗ · llama.cpp · prompt set v1

How we measure · The prompts

Link to this listing

Made or quantized this model? Paste one of these into your model card or README so people can find this listing.

Markdown

[Listed on Custom Clankers](https://customclankers.com/models/huihui-ai-huihui-qwen3-vl-8b-instruct-abliterated/)

HTML

<a href="https://customclankers.com/models/huihui-ai-huihui-qwen3-vl-8b-instruct-abliterated/">Listed on Custom Clankers</a>

Plain text links only: no image, no script, nothing loaded from this site.

Customize or build on this model

Listed from public GitHub data, not tested by us. Read the code and license before you run anything.

Linked in the model card

GitHub repositories the original model card links to.

Sumandora/remove-refusals-with-transformers ↗
Repo’s own description: Implements harmful/harmless refusal removal using pure HF Transformers
★ 2,312 · Last updated 2025-11-27 · License: Apache-2.0
ollama/ollama ↗
Repo’s own description: Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
★ 184,673 · Last updated 2026-10-24 · License: MIT

Projects that use this model

Public projects whose files name this exact model.

stephansturges/Tator ↗
Repo’s own description: Self-hosted dataset management and model-assisted annotation, with multi-user review, quality analysis and dataset pruning.
★ 92 · Last updated 2026-10-17 · License: not stated
Names this model in: tests/test_qwen_mlx_runtime.py ↗
senjinthedragon/Smart-Memory ↗
Repo’s own description: A SillyTavern extension providing a multi-tier memory and narrative context system for AI roleplay. Tracks character facts, relationship history, per-character knowledge and secrets, entity state…
★ 68 · Last updated 2026-10-01 · License: AGPL-3.0
Names this model in: README.md ↗
collbroGTR/comfyui-ideogram-autoprompter ↗
Repo’s own description: AI autoprompter for Ideogram 4's structured JSON caption format — local Qwen3-VL, Ollama or Gemini build the caption, then edit it on a visual bbox canvas.
★ 40 · Last updated 2026-06-15 · License: GPL-3.0
Names this model in: nodes/llm_local.py ↗

Tools to customize it further

Remove more refusals

Heretic ↗ ★ 36,421
Removes refusal behavior from transformer models automatically. It pairs directional ablation (abliteration) with an Optuna parameter search that tries to cut refusals while keeping the model close to the original, so you don't have to pick layers by hand.
OBLITERATUS ↗ ★ 8,946
A larger refusal-removal toolkit with several abliteration methods, usable locally or through a hosted Hugging Face Space. Check the telemetry setting before you run it: the README describes telemetry as opt-in, but on by default in the Spaces version.

Fine-tune it on your own data

Unsloth ↗ ★ 78,528
A desktop app and Python library for running and training models locally. Its README lists LoRA, QLoRA, full fine-tuning, DPO and GRPO, plus export to formats including GGUF, on Windows, Linux, WSL and macOS.
LlamaFactory ↗ ★ 75,713
Fine-tunes more than 100 language and vision-language models from a command line or a browser-based Web UI, aimed at training without writing code.

Make a smaller download (quantize)

llama.cpp ↗ ★ 132,260
Runs models locally and includes the tools behind most GGUF files. A conversion script turns Hugging Face weights into GGUF, and llama-quantize shrinks a high-precision GGUF to lower-bit formats, which can cost some accuracy.
MLX LM ↗ ★ 7,467
Apple's MLX package for running and fine-tuning models on Apple silicon. It can quantize models and upload them to the Hugging Face Hub, and supports LoRA and full fine-tuning.

Merge it with another model

mergekit ↗ ★ 7,414
Merges pretrained models with many merge methods. It works out-of-core, so merges can run on a CPU or with as little as 8 GB of VRAM, and it also handles LoRA extraction and mixture-of-experts merges.

Learn the techniques

LLM Course ↗ ★ 84,667
A free course with roadmaps and Colab notebooks. Its notebooks include one-click tools for merging with MergeKit, quantizing to GGUF, GPTQ, EXL2, AWQ and HQQ, and automatic abliteration with custom datasets.

Compare all tools →

Source provenance

Downloads: 26,689
Likes: 220
Metadata checked: 2026-10-25 · 05:05:23 UTC
First seen in index: 2026-10-20 · 21:59:08 UTC
Source revision: b47a0690b22eaf1d9a63874d967a03781c90f9cf

Source-card excerpt

Source model card excerpt: huihui-ai/Huihui-Qwen3-VL-8B-Instruct-abliterated This is an uncensored version of Qwen/Qwen3-VL-8B-Instruct created with abliteration (see remove-refusals-with-transformers to know more about it). It was only the text part

From huihui-ai/Huihui-Qwen3-VL-8B-Instruct-abliterated. Creator model-card excerpt, not an independent review.

Inspect before you run

Low refusal is not proof of correctness. Outputs can be inaccurate, harmful, or biased. Review the model card, license and files yourself. A license tag is not commercial-use advice. Parameter counts are source metadata; no hardware compatibility is implied.

This directory does not host or redistribute model weights. Refusal profiles are measured by us on a fixed prompt set; all other behavior claims are source-reported.

Collection rules & limitations ↗