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

juanml82/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-gguf

juanml82 Qwen 80B GGUF MoE 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/juanml82%2FHuihui-Qwen3-Next-80B-A3B-Instruct-abliterated-gguf"
Response includes
  • classification m8
  • files 5
  • benchmarks 11 entries
  • hub_downloads_all_time 8,145
  • author_summary 2 models
  • readme_text full
Try next
GET/models?author=juanml82
All 2 models by juanml82
 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
8K
331 last 30d - cooling
Likes
0
Model age
8mo ago
created 2026-01-16
Downloads over time
Now8.3K→from375↑2,108%
03K6K9.1K375 on Jan 148.3K on Oct 12JanMarMayJulSep
Jan 14 → Oct 12 · 79 snapshots · spans 271 days

Benchmarks

Benchmark Score Source
Entertainment 1.3 UGI
Hazardous 3.5 UGI
Natural Intelligence 27.28 UGI
Political lean -17.8% UGI
Sensitive-Info 20.22 UGI
SocPol 1.7 UGI
UGI 41.81 UGI
Willingness (10) 8.5 UGI
W10-Adherence 10 UGI
W10-Direct 7 UGI
Writing 42.37 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-Next-80B-A3B-Instruct-abliterated this lineage
juanml82/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-gguf↓ 331

Metadata

License
apache-2.0
Languages
en zh
Quantizations
IQ3
Tags
gguf en zh license:apache-2.0 endpoints_compatible region:us conversational

Related

Other by juanml82
Huihui-Qwen3-Next-80B-A3B-Thinking-abliterated-gguf 2K
Same base model
mradermacher/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-i1-GGUF 20K mradermacher/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-GGUF 2K coughmedicine/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-nvfp4 344 Eldadalbajob/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-mlx-3Bit 237 rayray916/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-mlx-3Bit-split-mtp 203
Total size
134 GB
Files
5
Quantizations
2
Registered
2026-08-22 13:56
Last updated on HF
2026-01-17 00:11
Source huggingface.co/juanml82/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated-gguf ↗ Base model huihui-ai/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated ↗

Files by quantization

IQ3 1 file 32.7 GB
qwen3-next-80b-a3b-instruct-IQ3_M.gguf 32.7 GB 562594c1 download
Auxiliary files 4 files 102 GB
qwen3-next-80b-a3b-instruct-q6_k.gguf 61.0 GB 92d285c0 download
qwen3-next-80b-a3b-instruct-mxfp4_moe.gguf 40.7 GB 19a36291 download
.gitattributes 1.71 KB cd8f823d download
README.md 819 B d12180f8 download

README current version from Hugging Face

license: apache-2.0
language:

  • en
  • zh
    base_model:
  • huihui-ai/Huihui-Qwen3-Next-80B-A3B-Instruct-abliterated

GGUF quants for Huihui-Qwen3-Next-80B-A3B-Thinking-abliterated

I've recreated them after the late December 2025 llama.cpp update which speeds up Qwen 3 Next, so these quants should perform better than the early quants for this model.
I've uploaded three quants:

iQ3_M – should fit (tight) in systems with 32gb of ram plus an 8-12gb gpu with ram offloading. Possibly lowest useful quant.

MXFP4_MOE – should work for systems with 32gb of ram plus a 16gb or more gpu.

Q6K – will work well with systems with 64gb of ram plus ram offloading. Quality is supposed to very almost indistinguishable from Q8

Enjoy!

README history 5 versions

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

  1. 2026-01-16Update README.mdc2f26a6819 B
    Loading...
  2. 2026-01-16Update README.md736764f889 B
    Loading...
  3. 2026-01-16Update README.md65ac0af1.1 KB
    Loading...
  4. 2026-01-16Update README.md5a8ec48118 B
    Loading...
  5. 2026-01-16initial commit7be884128 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