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

noctrex/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-GGUF

noctrex Qwen 30B GGUF MoE multimodal second-order 262K ctx
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/noctrex%2FHuihui-Qwen3-VL-30B-A3B-Thinking-abliterated-GGUF"
Response includes
  • classification m8
  • files 27
  • benchmarks 11 entries
  • hub_downloads_all_time 12,296
  • author_summary 44 models
  • readme_text full
Try next
GET/models?author=noctrex
All 44 models by noctrex
 GET/models?family=qwen&method=m8
Sibling qwen models · same method
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • author=noctrex (M8 quantization producer)
  • is_gguf=1
  • base_model='huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated' looks abliterated -> assume M1
How we label → About M8 →
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.

What is a refusal direction? →
Downloads · lifetime
12K
1K last 30d - cooling
Likes
3
Model age
11mo ago
created 2025-11-07
Downloads over time
Now12.6K→from3.6K↑247%
3.2K6.6K10.1K13.5K3.6K on Nov 5, 202512.6K on Oct 11Nov '25JanMarMayJulSep
Nov 5, 2025 → Oct 11 · 89 snapshots · spans 340 days

Benchmarks

Benchmark Score Source
Entertainment 1 UGI
Hazardous 1.2 UGI
Natural Intelligence 14.12 UGI
Political lean -10.9% UGI
Sensitive-Info 10.42 UGI
SocPol 1 UGI
UGI 34.44 UGI
Willingness (10) 8.2 UGI
W10-Adherence 7.5 UGI
W10-Direct 9 UGI
Writing 22.11 UGI

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-VL-30B-A3B-Thinking-abliterated↓ 144 this lineage
noctrex/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-GGUF↓ 1,188

Metadata

Quantizations
IQ2 IQ3 IQ4 Q2_K Q3_K Q4_K Q5_K Q6_K Q8_0
Tags
gguf image-text-to-text base_model:huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated base_model:quantized:huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated endpoints_compatible region:us imatrix conversational

Related

Other by noctrex
Huihui-Qwen3-VL-4B-Instruct-abliterated-GGUF 28K Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF 15K Huihui-Qwen3-VL-8B-Instruct-abliterated-GGUF 9K gemma-4-26B-A4B-it-uncensored-heretic-MXFP4_MOE-GGUF 3K Huihui-Qwen3-VL-2B-Instruct-abliterated-GGUF 3K
Same base model
mradermacher/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-i1-GGUF 3K mradermacher/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-GGUF 590 IceHibiki/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q4_K_M-GGUF 37
Total size
317 GB
Files
27
Quantizations
13
Registered
2026-08-22 13:56
Last updated on HF
2025-11-07 17:08
Source huggingface.co/noctrex/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-GGUF ↗ Base model huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated →

Files by quantization

Q8_0 1 file 30.3 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q8_0.gguf 30.3 GB c2b621d8 download
Q6_K 1 file 23.4 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q6_K.gguf 23.4 GB f4926ac5 download
Q5_K 2 files 39.9 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q5_K_M.gguf 20.2 GB a693002f download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q5_K_S.gguf 19.6 GB 7bfdfd4d download
Q4_K 2 files 33.5 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q4_K_M.gguf 17.3 GB 187643be download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q4_K_S.gguf 16.3 GB 123265e5 download
IQ4 2 files 31.4 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ4_NL.gguf 16.1 GB 1559e4be download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ4_XS.gguf 15.2 GB eaee87b5 download
Q3_K 3 files 40.9 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q3_K_L.gguf 14.8 GB f9f32c57 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q3_K_M.gguf 13.7 GB 5b80e5b3 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q3_K_S.gguf 12.4 GB 3d66e1d6 download
IQ3 4 files 47.7 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ3_M.gguf 12.6 GB 2e0e3622 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ3_S.gguf 12.4 GB 819bbf60 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ3_XS.gguf 11.7 GB f975a529 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ3_XXS.gguf 11.0 GB e91556b0 download
Q2_K 2 files 20.3 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q2_K.gguf 10.5 GB 907140d8 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-Q2_K_S.gguf 9.80 GB f408adfc download
IQ2 3 files 26.6 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ2_M.gguf 9.47 GB d68a4460 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ2_S.gguf 8.65 GB a9f58edf download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-IQ2_XS.gguf 8.45 GB 64855603 download
F32 1 file 2.01 GB
mmproj-F32.gguf 2.01 GB 5e9feee7 download
BF16 1 file 1.01 GB
mmproj-BF16.gguf 1.01 GB d74ff447 download
F16 1 file 1.01 GB
mmproj-F16.gguf 1.01 GB dfd20e1c download
Auxiliary files 4 files 23.5 GB
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-MXFP4_MOE.gguf 15.9 GB 8958e8e3 download
Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated-TQ2_0.gguf 7.63 GB af9a3f46 download
.gitattributes 3.63 KB 139c113b download
README.md 526 B 8c068b55 download

README current version from Hugging Face

pipeline_tag: image-text-to-text
base_model:

  • huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated

These are quantizations of the model Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated

The imatrix from unsloth has been used for these.

Original model: https://huggingface.co/huihui-ai/Huihui-Qwen3-VL-30B-A3B-Thinking-abliterated

Download the latest llama.cpp to use them.

Try to use the best quality you can run.
For the mmproj, try to use the F32 version as it will produce the best results.
F32 > BF16 > F16.

README history 3 versions

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

  1. 2025-11-07Add files using upload-large-folder toolede88c9526 B
    Loading...
  2. 2025-11-07Add files using upload-large-folder tool20d97db382 B
    Loading...
  3. 2025-11-07Add files using upload-large-folder tool1f44ae2331 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration