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

nbeerbower/Huihui-Qwen3.5-9B-abliterated-TIES

nbeerbower Qwen 9.7B multimodal
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/nbeerbower%2FHuihui-Qwen3.5-9B-abliterated-TIES"
Response includes
  • classification m5
  • files 14
  • hub_downloads_all_time 123
  • author_summary 27 models
  • readme_text full
Try next
GET/models?author=nbeerbower
All 27 models by nbeerbower
 GET/models?family=qwen&method=m5
Sibling qwen models · same method
 GET/methods/m5
What "m5" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M5
Primary method

Mergekit merge

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.
  • merge tag / mergekit / dare-ties in tags or name
  • no unusual architecture pattern (regular merge)
  • abliterated marker present
How we label → About M5 →
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
123
28 last 30d - stable
Likes
1
Descendants
6
in 3 direct forks
Model age
6mo ago
created 2026-03-24
Downloads over time
Now135→from14↑864%
85410114714 on Mar 25135 on Oct 11MarAprMayJunJulAugSepOct
Mar 25 → Oct 11 · 68 snapshots · spans 200 days

Genealogy 3 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.

Qwen/Qwen3.5-9B-Base this lineage
nbeerbower/Huihui-Qwen3.5-9B-abliterated-TIES↓ 28 3 forks
nbeerbower/Huihui-Qwen3.5-9B-abliterated-TIES-Hemlock-SFT↓ 32 3 forks
mradermacher/Huihui-Qwen3.5-9B-abliterated-TIES-Hemlock-SFT-i1-GGUF↓ 1,185
mradermacher/Huihui-Qwen3.5-9B-abliterated-TIES-Hemlock-SFT-GGUF↓ 371
fairy322/Huihui-Qwen3.5-9B-abliterated-TIES-Hemlock-SFT-GGUF↓ 363
mradermacher/Huihui-Qwen3.5-9B-abliterated-TIES-i1-GGUF↓ 1,179
mradermacher/Huihui-Qwen3.5-9B-abliterated-TIES-GGUF↓ 624

Metadata

Tags
transformers safetensors qwen3_5 image-text-to-text mergekit merge conversational arxiv:2306.01708 base_model:Qwen/Qwen3.5-9B-Base base_model:merge:Qwen/Qwen3.5-9B-Base base_model:huihui-ai/Huihui-Qwen3.5-9B-abliterated base_model:merge:huihui-ai/Huihui-Qwen3.5-9B-abliterated

Related

Other by nbeerbower
Huihui-Qwen3.5-27B-abliterated-Athanorlite-ORPO-v2-GGUF 191 llama-3-spicy-abliterated-stella-8B 46 EVA-abliterated-Qwen2.5-7B 46 Huihui-Qwen3.5-9B-abliterated-Grimoire-SimPO 46 Qwen3-8B-abliterated-TIES 45
Same base model
GitMylo/Qwen3.5-9B-Uncensored-HauhauCS-Aggressive-safetensors 1K MethodWhite/Qwen3.5-9B-Abliterated-HSAQ-v2 814 MethodWhite/Qwen3.5-9B-Abliterated-HSAQ 92 MiguelMendez101/Qwen3.5-9B-uncensored-ara 69 bowmanslayer/Qwen3.5-9B-Uncensored 58
Total size
18.0 GB
Files
14
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-03-24 18:19
Source huggingface.co/nbeerbower/Huihui-Qwen3.5-9B-abliterated-TIES ↗ Base model Qwen/Qwen3.5-9B-Base ↗

Files by quantization

Auxiliary files 14 files 18.1 GB
model-00002-of-00004.safetensors 4.65 GB d847cb47 download
model-00003-of-00004.safetensors 4.61 GB 93ebd27c download
model-00001-of-00004.safetensors 4.60 GB 4f3a56c3 download
model-00004-of-00004.safetensors 2.88 GB 9f9b805c download
model-vlm.safetensors 1.30 GB 21cbc389 download
tokenizer.json 12.2 MB fe000e3e download
model.safetensors.index.json 65.0 KB e88f8a04 download
tokenizer_config.json 16.3 KB ae8d254e download
config.json 3.05 KB 273ce437 download
README.md 2.85 KB fe09f422 download
.gitattributes 1.53 KB 52373fe2 download
mergekit_config.yml 465 B 434a55b0 download
preprocessor_config.json 390 B 2ea84a43 download
video_preprocessor_config.json 386 B 37900b3f download

README current version from Hugging Face

base_model:

  • Qwen/Qwen3.5-9B-Base
  • nbeerbower/Huihui-Qwen3.5-9B-abliterated-Grimoire-SFT
  • nbeerbower/Huihui-Qwen3.5-9B-abliterated-Grimoire-ORPO
  • huihui-ai/Huihui-Qwen3.5-9B-abliterated
    library_name: transformers
    tags:
  • mergekit
  • merge

Huihui-Qwen3.5-9B-abliterated-TIES

This is a merge of pre-trained language models created using a custom fork of mergekit with Qwen3.5 architecture support.

Merge Details

Merge Method

This model was merged using the TIES merge method using Qwen/Qwen3.5-9B-Base as a base.

Models Merged

The following models were included in the merge:

Mergekit Changes for Qwen3.5

Qwen3.5 uses a hybrid attention architecture (3:1 linear/full attention layers) that mergekit does not yet support upstream. The following changes were made to enable this merge:

  1. New architecture definitions (qwen3_5.json, qwen3_5_text.json) - Defines tensor mappings for both Qwen3_5ForConditionalGeneration (VLM) and Qwen3_5ForCausalLM (text-only) architectures. The hybrid self_attn and linear_attn layer weights are marked as optional since they only appear on specific layers (full attention every 4th layer, linear attention on the rest).

  2. Config key fallback (mergekit/common.py) - Added a fallback in get_config_value so that nested config keys like text_config.num_hidden_layers gracefully resolve to num_hidden_layers on text-only model configs. This enables cross-architecture merges between VLM and text-only Qwen3.5 variants.

  3. Vision weight grafting - The SFT and ORPO fine-tuned models are text-only (Qwen3_5ForCausalLM) and lack the vision encoder. After the text merge, the vision encoder (model.visual.*) and MTP head (mtp.*) weights were grafted from the base VLM model (Qwen/Qwen3.5-9B-Base) to produce a complete VLM.

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: huihui-ai/Huihui-Qwen3.5-9B-abliterated
    parameters:
      weight: 0.4
      density: 0.6
  - model: nbeerbower/Huihui-Qwen3.5-9B-abliterated-Grimoire-SFT
    parameters:
      weight: 0.3
      density: 0.6
  - model: nbeerbower/Huihui-Qwen3.5-9B-abliterated-Grimoire-ORPO
    parameters:
      weight: 0.5
      density: 0.6
merge_method: ties
base_model: Qwen/Qwen3.5-9B-Base
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16

README history 1 version

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

  1. 2026-03-24Upload folder using huggingface_hub05785262.8 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.

Get the app Browse the catalog
Open in Abliteration