← back to catalog · registered 2026-11-15 05:03

tmthang86/ThinkingCap-Qwen3.8-27B-abliterated-mlx-4bit

tmthang86 27B 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/tmthang86%2FThinkingCap-Qwen3.8-27B-abliterated-mlx-4bit"
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
98
Likes
0
Model age
1w ago
created 2026-11-15
Downloads over time
Now 98 → from 0 ↑ 0%
03672108 0 on Nov 15 98 on Nov 22 Nov
Nov 15 → Nov 22 · 8 snapshots · spans 7 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
other
Languages
en multilingual
Tags
mlx safetensors qwen3_5 qwen3_8 thinkingcap abliterated uncensored text-generation conversational en multilingual base_model:IstroSec/ThinkingCap-Qwen3.8-27B-abliterated

Related

Total size
14.1 GB
Files
13
Quantizations
1
Registered
2026-11-15 05:03
Last updated on HF
2026-11-15 04:53

Files by quantization

Auxiliary files 13 files 14.1 GB
model-00002-of-00003.safetensors 4.99 GB 8b52afc9 download
model-00001-of-00003.safetensors 4.96 GB bd68dc91 download
model-00003-of-00003.safetensors 4.14 GB e1a0523e download
tokenizer.json 19.1 MB a5cd9732 download
model.safetensors.index.json 185 KB 8cabf784 download
chat_template.jinja 8.74 KB c0c686f9 download
config.json 4.01 KB fa0b39f1 download
README.md 1.81 KB 84f4972f download
.gitattributes 1.53 KB 52373fe2 download
LICENSE 1.23 KB c6b9bd63 download
tokenizer_config.json 1.20 KB 159b5c47 download
NOTICE 757 B ced5b77f download
generation_config.json 214 B c53835dc download

README current version from Hugging Face


license: other
license_name: polyform-small-business-1.0.0
license_link: LICENSE
base_model: IstroSec/ThinkingCap-Qwen3.8-27B-abliterated
base_model_relation: quantized
library_name: mlx
pipeline_tag: text-generation
language:

  • en
  • multilingual
    tags:
  • mlx
  • qwen3_5
  • qwen3_8
  • thinkingcap
  • abliterated
  • uncensored

ThinkingCap-Qwen3.8-27B-abliterated, MLX 4-bit

An MLX 4-bit build (4.5 bits per weight, about 14 GB) of
IstroSec/ThinkingCap-Qwen3.8-27B-abliterated,
the uncensored variant of bottlecapai/ThinkingCap-Qwen3.8-27B
(a Qwen3.8-27B fine-tune that spends fewer thinking tokens).

Converted with mlx-lm 0.32.2, standard uniform 4-bit quantization, no changes to the weights
or the chat template:

mlx_lm.convert --hf-path IstroSec/ThinkingCap-Qwen3.8-27B-abliterated --mlx-path out-4bit -q --q-bits 4

Text generation through mlx-lm only; the vision input and the MTP draft head of the source
checkpoint were not tested in this build.

Use

pip install "mlx-lm>=0.32.2"
mlx_lm.generate --model tmthang86/ThinkingCap-Qwen3.8-27B-abliterated-mlx-4bit --max-tokens 3000 --prompt "..."

It is a thinking model: leave room for the reasoning before the answer.

Checked

Loaded on an Apple Silicon Mac with 48 GB: about 33 tokens per second, peak memory 15.8 GB,
valid JSON on a short Vietnamese contradiction-finding prompt. No benchmark was run on this build.

Licence

PolyForm Small Business License 1.0.0, with BottleCap AI's additional personal-use permission,
as in the source repository. LICENSE and NOTICE are copied unchanged. Check them before any
commercial use. The Qwen base model is by Alibaba Cloud / Qwen Team (Apache 2.0).

README history 1 version

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

  1. 2026-11-15 MLX 4-bit build of IstroSec/ThinkingCap-Qwen3.8-27B-abliterated (mlx-lm 0.32.2) 5e943fe 1.8 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