← back to catalog · registered 2026-11-11 00:03

TheShellMaster/cypher-Huihui-Qwen3.5-2B-abliterated

TheShellMaster 2B GGUF 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/TheShellMaster%2Fcypher-Huihui-Qwen3.5-2B-abliterated"
Response includes
  • classification m8
  • files 3
  • author_summary 15 models
  • readme_text full
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
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
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 · 30-day
180
Likes
0
Model age
1w ago
created 2026-11-10
Downloads over time
Now 180 → from 0 ↑ 0%
066132198 0 on Nov 11 180 on Nov 22 Nov
Nov 11 → Nov 22 · 12 snapshots · spans 11 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
Languages
fr en
Quantizations
Q5_K
Tags
gguf llama.cpp q5_k_m abliterated uncensored texte text-generation fr en base_model:huihui-ai/Huihui-Qwen3.5-2B-abliterated base_model:quantized:huihui-ai/Huihui-Qwen3.5-2B-abliterated license:apache-2.0

Related

Total size
1.33 GB
Files
3
Quantizations
2
Registered
2026-11-11 00:03
Last updated on HF
2026-11-10 23:21

Files by quantization

Q5_K 1 file 1.33 GB
Huihui-Qwen3.5-2B-abliterated.Q5_K_M.gguf 1.33 GB f6b5287d download
Auxiliary files 2 files 4.34 KB
README.md 2.78 KB a8f35d91 download
.gitattributes 1.56 KB 558a0d29 download

README current version from Hugging Face


license: apache-2.0
base_model: huihui-ai/Huihui-Qwen3.5-2B-abliterated
language:

  • fr
  • en
    tags:
  • gguf
  • llama.cpp
  • q5_k_m
  • abliterated
  • uncensored
  • texte
    pipeline_tag: text-generation

cypher-Huihui-Qwen3.5-2B-abliterated

Miroir public et non modifié du fichier Huihui-Qwen3.5-2B-abliterated.Q5_K_M.gguf (Q5_K_M, 1,42 Go) publié par
mradermacher/Huihui-Qwen3.5-2B-abliterated-GGUF.

Aucune re-quantification, aucun fine-tuning, aucun changement de template : les poids
sont bit pour bit identiques à la source.

sha256  5b3629a94dd157d1e3e7cd339c4b53bea0e47822b5dcfc307b6b68737100cd7f
taille  1424769472 octets

Ce que c'est

Type texte
Paramètres 2,3 B
Architecture qwen35
Contexte natif 262 144 tokens
Quantification Q5_K_M (~5,4 bits/poids)
Poids sur disque 1,42 Go
Usage visé 2 vCPU / Space gratuit
Licence apache-2.0

Généalogie

Étape Dépôt
Modèle racine Qwen/Qwen3.5-2B
Abliteration (fp16) huihui-ai/Huihui-Qwen3.5-2B-abliterated
Quantification GGUF mradermacher/Huihui-Qwen3.5-2B-abliterated-GGUF
Miroir (ce dépôt) TheShellMaster/cypher-Huihui-Qwen3.5-2B-abliterated

Utilisation

# Téléchargement
hf download TheShellMaster/cypher-Huihui-Qwen3.5-2B-abliterated Huihui-Qwen3.5-2B-abliterated.Q5_K_M.gguf --local-dir .

# llama.cpp
llama-cli -m Huihui-Qwen3.5-2B-abliterated.Q5_K_M.gguf -p "Bonjour" -c 8192 -ngl 99

Avec Ollama, créez un Modelfile :

FROM ./Huihui-Qwen3.5-2B-abliterated.Q5_K_M.gguf
PARAMETER num_ctx 8192
PARAMETER temperature 0.7

puis ollama create qwen35-2b -f Modelfile.

Quantification Q5_K_M

K-quant mixte par blocs de 256 poids : environ 5,3 à 5,7 bits par poids en moyenne
(couches d'attention plus précises que les couches FFN, échelles incluses).
Repères : Q4_K_M ≈ 4,8 · Q5_K_M ≈ 5,4 · Q6_K ≈ 6,6 · Q8_0 ≈ 8,5 bits/poids.

Attribution

Ce dépôt n'est qu'un point de distribution redondant. Tout le crédit revient aux
auteurs ci-dessus ; en cas de divergence, la source fait foi.

README history 1 version

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

  1. 2026-11-10 Upload README.md with huggingface_hub 269f707 2.8 KB
    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.

Open in Abliteration