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

lukaskremla/Huihui-Nex-N2-mini-abliterated-mlx-6Bit

lukaskremla MoE multimodal second-order
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/lukaskremla%2FHuihui-Nex-N2-mini-abliterated-mlx-6Bit"
Response includes
  • classification m1
  • files 13
  • author_summary 3 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
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

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
16
↑ 459% in 90 days
Likes
0
Model age
3mo ago
created 2026-06-17
Downloads over time
Now229→from41↑459%
3210417624841 on Jun 17229 on Aug 30229 on Aug 26JunJulAug
Jun 17 → Aug 30 · 14 snapshots · spans 74 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.

Metadata

License
apache-2.0
Tags
transformers safetensors qwen3_5_moe image-text-to-text abliterated uncensored mlx mlx-my-repo conversational base_model:huihui-ai/Huihui-Nex-N2-mini-abliterated base_model:quantized:huihui-ai/Huihui-Nex-N2-mini-abliterated license:apache-2.0

Related

Total size
26.2 GB
Files
13
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-06-17 19:16

Files by quantization

Auxiliary files 13 files 26.3 GB
model-00003-of-00006.safetensors 4.89 GB 2658d0f8 download
model-00004-of-00006.safetensors 4.89 GB 667102ab download
model-00002-of-00006.safetensors 4.87 GB 61f5528d download
model-00001-of-00006.safetensors 4.87 GB ad156dd6 download
model-00005-of-00006.safetensors 4.86 GB c55c1182 download
model-00006-of-00006.safetensors 1.86 GB 2e98a34c download
tokenizer.json 19.1 MB 87a7830d download
model.safetensors.index.json 183 KB d155a813 download
config.json 22.6 KB 382552cf download
chat_template.jinja 7.57 KB fa6e2772 download
.gitattributes 1.53 KB 52373fe2 download
tokenizer_config.json 1.15 KB 040f6e4b download
README.md 1.05 KB 52173676 download

README current version from Hugging Face


library_name: transformers
license: apache-2.0
pipeline_tag: image-text-to-text
base_model: huihui-ai/Huihui-Nex-N2-mini-abliterated
tags:

  • abliterated
  • uncensored
  • mlx
  • mlx-my-repo

lukaskremla/Huihui-Nex-N2-mini-abliterated-mlx-6Bit

The Model lukaskremla/Huihui-Nex-N2-mini-abliterated-mlx-6Bit was converted to MLX format from huihui-ai/Huihui-Nex-N2-mini-abliterated using mlx-lm version 0.31.2.

Use with mlx

pip install mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("lukaskremla/Huihui-Nex-N2-mini-abliterated-mlx-6Bit")

prompt="hello"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
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.

Open in Abliteration