Downloads · 30 days
527
8% of all-time downloads
dealignai/Qwen3.5-VL-397B-A17B-UNCENSORED-JANG_1L
Qwen3.5-VL-397B-A17B-UNCENSORED-JANG_1L is a image-text-to-text model from dealignai. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for mlx. The card lists the license as apache-2.0.
Important: This model uses the JANG quantization format — the GGUF equivalent for MLX on Apple Silicon. Currently only supported by MLX Studio and the jang-tools Python package.
Downloads · 30 days
527
8% of all-time downloads
All-time downloads
7K
Public
Parameters
429B
121 GB on disk
Likes
5
Public
Click a slice to open those files.
.safetensors121 GB · 100%
How the weights are stored.
U32428B · 100%
From the Hugging Face model README
Important: This model uses the JANG quantization format — the GGUF equivalent for MLX on Apple Silicon. Currently only supported by MLX Studio and the
jang-toolsPython package.
JANG mixed-precision · CRACK abliterated · Vision-Language · No guardrails · 112 GB
<a href="https://ko-fi.com/jangq"><img src="https://img.shields.io/badge/Ko--fi-Support_Development-FF5E5B?logo=ko-fi&logoColor=white&style=for-the-badge" alt="Ko-fi"></a>
</div>This is Qwen 3.5 VL 397B — a 397B parameter hybrid SSM/Attention Mixture-of-Experts model with 512 experts (10 active per token), GatedDeltaNet SSM + full attention layers, and built-in vision.
It has been:
| Architecture | Qwen 3.5 VL MoE — 397B total, ~17B active, 512 experts, hybrid SSM/FA |
| Quantization | JANG_1L (8/2-bit mixed, 2.13 avg) — 112 GB |
| Abliteration | CRACK — weight-level surgery |
| HarmBench | 96.2% (308/320) |
| Compliance | 8/8 |
| Speed | 33 tok/s (M3 Ultra 256GB) |
| Vision | Yes — via MLX Studio / vMLX |
| Thinking | ON/OFF supported |
| Fits on | 128 GB+ Macs (tight) / 256 GB Macs (comfortable) |
308/320 (96.2%) — tested with v2 matcher
| Category | Score | |
|---|---|---|
| Copyright | 80/80 | 100% |
| Misinformation / Disinfo | 54/54 | 100% |
| Chemical / Biological | 41/42 | 98% |
| Cybercrime / Intrusion | 50/52 | 96% |
| Illegal | 49/53 | 92% |
| Harmful | 16/18 | 89% |
| Harassment / Bullying | 18/21 | 86% |
185/208 (88.9%) — 208 questions across 13 subjects, thinking recovery on failures
| CRACK | Base JANG_1L | Delta | |
|---|---|---|---|
| MMLU | 88.9% | 87.0% | +1.9% |
| Speed | 33 tok/s | 36 tok/s | -8% |
| HarmBench | 96.2% | 0% | +96.2% |
| Subject | CRACK | /16 | Type |
|---|---|---|---|
| Professional Medicine | 16/16 | 100% | HARD |
| HS Biology | 16/16 | 100% | BASE |
| World Religions | 16/16 | 100% | BASE |
| College Physics | 15/16 | 94% | HARD |
| Conceptual Physics | 15/16 | 94% | HARD |
| HS Geography | 15/16 | 94% | BASE |
| Electrical Engineering | 14/16 | 88% | HARD |
| College CS | 13/16 | 81% | HARD |
| Machine Learning | 13/16 | 81% | HARD |
| Abstract Algebra | 12/16 | 75% | HARD |
| HS Mathematics | 12/16 | 75% | HARD |
| Formal Logic | 11/16 | 69% | HARD |
| College Mathematics | 11/16 | 69% | HARD |
| Total | 185/208 | 88.9% |
Surgery improved reasoning — safety guardrails were interfering with mathematical problem-solving.
pip install "jang[mlx]"
from jang_tools.loader import load_jang_model
from mlx_lm import generate
model, tokenizer = load_jang_model("dealignai/Qwen3.5-397B-A17B-JANG_1L-CRACK")
messages = [{"role": "user", "content": "Your prompt here"}]
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True, tokenize=False)
response = generate(model, tokenizer, prompt=prompt, max_tokens=2000)
print(response)
Thinking is ON by default. To disable:
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True,
enable_thinking=False, tokenize=False)
JANG (Jang Adaptive N-bit Grading) is a mixed-precision quantization format for Apple Silicon — the GGUF equivalent for MLX.
CRACK (Controlled Refusal Ablation via Calibrated Knockouts) removes safety alignment from LLMs at the weight level using per-layer projected vectors from structurally-mirrored prompt pairs.
This model is provided for research and educational purposes. The creators are not responsible for any misuse. By downloading this model, you agree to use it responsibly and in compliance with applicable laws.
| 항목 | 내용 |
|---|---|
| 크기 | 112 GB |
| HarmBench | 96.2% (308/320) |
| 속도 | 33 tok/s (M3 Ultra) |
| 비전 | 지원 (MLX Studio / vMLX) |
| 최소 요구사양 | 128 GB 메모리 Mac |
pip install "jang[mlx]"
GitHub · HuggingFace · MLX Studio · Ko-fi · X @dealignai