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

Tipur/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF

Tipur 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/Tipur%2FHuihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF"
Response includes
  • classification m8
  • files 27
  • hub_downloads_all_time 6,228
  • author_summary 1 models
  • readme_text full
Try next
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
MEDIUM
Inherited from base model
M1 Direct removal →
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=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
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
6K
591 last 30d - cooling
Likes
0
Model age
4mo ago
created 2026-05-15
Downloads over time
Now6.4K→from1.5K↑332%
1.2K3.1K5K6.9K1.5K on May 206.4K on Oct 11MayJunJulAugSepOct
May 20 → Oct 11 · 60 snapshots · spans 144 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-VL-30B-A3B-Instruct-abliterated↓ 3,189 this lineage
Tipur/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF↓ 591

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-Instruct-abliterated base_model:quantized:huihui-ai/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated endpoints_compatible region:us imatrix conversational

Related

Same base model
noctrex/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF 15K mradermacher/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF 12K mradermacher/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-i1-GGUF 8K McG-221/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q8_0-GGUF 547 MLXBits/huihui-qwen3-vl-30b-abliterated-4bit 99
Total size
310 GB
Files
27
Quantizations
13
Registered
2026-08-22 13:56
Last updated on HF
2026-05-15 17:15
Source huggingface.co/Tipur/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF ↗ Base model huihui-ai/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated →

Files by quantization

Q8_0 2 files 30.9 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q8_0.gguf 30.3 GB f30a0775 download
mmproj-Q8_0.gguf 679 MB 4dcdbf32 download
Q6_K 1 file 23.4 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q6_K.gguf 23.4 GB 9d9ab9ec download
Q5_K 2 files 39.9 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q5_K_M.gguf 20.2 GB c89eef0f download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q5_K_S.gguf 19.6 GB 55527573 download
Q4_K 2 files 33.5 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q4_K_M.gguf 17.3 GB 1e94a651 download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q4_K_S.gguf 16.3 GB 9cc17ccf download
IQ4 2 files 31.4 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ4_NL.gguf 16.1 GB defbb513 download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ4_XS.gguf 15.2 GB b6d6c49e download
Q3_K 3 files 40.9 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q3_K_L.gguf 14.8 GB 0d5a48eb download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q3_K_M.gguf 13.7 GB 0970f45d download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q3_K_S.gguf 12.4 GB 5f858ca4 download
IQ3 4 files 47.7 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ3_M.gguf 12.6 GB 1bbf39fe download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ3_S.gguf 12.4 GB 03ec16b8 download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ3_XS.gguf 11.7 GB 958b8207 download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ3_XXS.gguf 11.0 GB cee5d000 download
Q2_K 2 files 20.3 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q2_K.gguf 10.5 GB cf70724c download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-Q2_K_S.gguf 9.80 GB 1047b967 download
IQ2 3 files 26.6 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ2_M.gguf 9.47 GB e4bf52ab download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ2_S.gguf 8.65 GB a3139a8b download
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-IQ2_XS.gguf 8.45 GB 096a3384 download
F32 1 file 2.01 GB
mmproj-F32.gguf 2.01 GB b5aaaf28 download
BF16 1 file 1.01 GB
mmproj-BF16.gguf 1.01 GB 567b6826 download
F16 1 file 1.01 GB
mmproj-F16.gguf 1.01 GB 4bfd6558 download
Auxiliary files 3 files 15.9 GB
Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-MXFP4_MOE.gguf 15.9 GB 5f458db6 download
.gitattributes 3.59 KB 81d92ae5 download
README.md 526 B f7664857 download

README current version from Hugging Face

pipeline_tag: image-text-to-text
base_model:

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

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

The imatrix from unsloth has been used for these.

Original model: https://huggingface.co/huihui-ai/Huihui-Qwen3-VL-30B-A3B-Instruct-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 1 version

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

  1. 2026-05-15Duplicate from noctrex/Huihui-Qwen3-VL-30B-A3B-Instruct-abliterated-GGUF53ed142526 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