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

vanch007/Huihui-Qwen3.6-27B-abliterated-mlx-bf16

Name: vanch007/Huihui-Qwen3.6-27B-abliterated-mlx-bf16
Author: vanch007

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/vanch007%2FHuihui-Qwen3.6-27B-abliterated-mlx-bf16"

Response includes

classification m1
files 24
hub_downloads_all_time 1,419
author_summary 27 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

25 last 30d - cooling

Likes

Model age

5mo ago

created 2026-04-23

Downloads over time

Apr 22 → Oct 11 · 64 snapshots · spans 172 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.

huihui-ai/Huihui-Qwen3.6-27B-abliterated↓ 4,286 this lineage

vanch007/Huihui-Qwen3.6-27B-abliterated-mlx-bf16↓ 25

Variants by this author 2 formats · 104 downloads combined

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

vanch007/Huihui-Qwen3.6-27B-abliterated-mlx-nvfp4 NVFP4 79
vanch007/Huihui-Qwen3.6-27B-abliterated-mlx-bf16 (this) BF16 / F16 25

Metadata

Files by quantization

Auxiliary files 24 files 51.0 GB

model-00009-of-00011.safetensors	4.98 GB	c418689c	download
model-00005-of-00011.safetensors	4.98 GB	7d19a97a	download
model-00004-of-00011.safetensors	4.96 GB	1c235208	download
model-00006-of-00011.safetensors	4.96 GB	5ae8ed9a	download
model-00003-of-00011.safetensors	4.96 GB	57d59a2f	download
model-00007-of-00011.safetensors	4.96 GB	50ead314	download
model-00008-of-00011.safetensors	4.96 GB	a2499022	download
model-00002-of-00011.safetensors	4.96 GB	2cae2d4f	download
model-00001-of-00011.safetensors	4.87 GB	f73b1b30	download
model-00010-of-00011.safetensors	4.03 GB	a6b5521b	download
model-00011-of-00011.safetensors	2.37 GB	32f93d8c	download
tokenizer.json	19.1 MB	87a7830d	download
vocab.json	6.41 MB	0aa0ce06	download
model.safetensors.index.json	113 KB	3425801b	download
chat_template.jinja	7.58 KB	a8755d82	download
config.json	4.27 KB	01aae98e	download
.gitattributes	1.53 KB	52373fe2	download
README.md	1.28 KB	ce427eb5	download
processor_config.json	1.27 KB	7ad6acdf	download
tokenizer_config.json	1.11 KB	a068e246	download
preprocessor_config.json	390 B	2ea84a43	download
video_preprocessor_config.json	385 B	3ba673a5	download
generation_config.json	202 B	023756cf	download
configuration.json	51.0 B	3a6d4256	download

README current version from Hugging Face

library_name: mlx-vlm
pipeline_tag: image-text-to-text
base_model: huihui-ai/Huihui-Qwen3.6-27B-abliterated

tags:

mlx
mlx-vlm
qwen3.6
multimodal
vision-language
abliterated

Huihui-Qwen3.6-27B-abliterated-mlx-bf16

Local MLX-VLM conversion of huihui-ai/Huihui-Qwen3.6-27B-abliterated.

Overview

Format: MLX-VLM
Precision: bf16
Size: 51G
q_mode: bf16
q_bits: bf16
Source model type: Qwen3_5ForConditionalGeneration
Source pipeline: image-text-to-text
Intended runtime: mlx-vlm, LM Studio

Validation

text generation smoke test: passed
image generation path smoke test: passed

Quantization

Quantization: none (bf16 export)

Test Results (2026-04-23)

Test runner: mlx_vlm.generate

Text prompt:
你好，请用一句自然中文回应。

Image prompt:
请用一句中文描述这张图片。

Image asset:
64x64 pure red square

Observed results:

Case	Prompt TPS	Generation TPS	Peak Memory
Text	11.309	6.471	54.861 GB
Image	26.533	6.165	55.004 GB

Notes:

Both text and image tests completed successfully.
The current mlx_vlm runtime still emits <think> content even with --processor-kwargs '{"enable_thinking": false}'.

README history 1 version

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

2026-04-23Add files using upload-large-folder tool2ff39e71.3 KB
Loading...