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.
Qwen 3.5 VL 4B — JANG_4S + CRACK
JANG mixed-precision · CRACK abliterated · Vision-Language · No guardrails · 3 GB
What Is This?
This is Qwen 3.5 VL 4B — a 4B parameter dense hybrid SSM/Attention model with built-in vision capabilities. The smallest model in the Qwen 3.5 family that still delivers solid performance.
It has been:
- JANG quantized — JANG_4S profile (6-bit attention, 4-bit MLP) — 3 GB
- CRACK abliterated — permanent weight-level removal of safety refusal
| Architecture | Qwen 3.5 VL Dense — 4B params, hybrid SSM/FA, 32 layers |
| Quantization | JANG_4S (6/4-bit mixed) — 3 GB |
| Abliteration | CRACK — novel weight surgery |
| HarmBench | 91.2% (292/320) |
| MMLU | 63.1% (base: 56.9%, +6.2% improvement) |
| Compliance | 8/8 |
| Speed | 134 tok/s (M4 Max) |
| Vision | Yes — via MLX Studio / vMLX |
| Thinking | ON/OFF supported |
| Fits on | 8 GB+ Macs |
JANG vs MLX Uniform Quantization
| Model | MMLU | Size | Speed | Notes |
|---|---|---|---|---|
| JANG_4S + CRACK | 63.1% | 3 GB | 134 tok/s | This model |
| JANG_4S (base) | 67.5% | 3 GB | 134 tok/s | Unmodified JANG |
| MLX 4-bit | 67.0% | 2.2 GB | ~100 tok/s | Uniform quant |
HarmBench Results
292/320 (91.2%) — tested with enable_thinking=false, temperature=1.0
| Category | Score | |
|---|---|---|
| Chemical / Biological | 41/42 | 98% |
| Cybercrime / Intrusion | 51/52 | 98% |
| Misinformation / Disinfo | 50/54 | 93% |
| Illegal | 49/53 | 92% |
| Harmful | 16/18 | 89% |
| Copyright | 68/80 | 85% |
| Harassment / Bullying | 17/21 | 81% |
MMLU Results
Surgery improved the model's reasoning — safety guardrails were interfering with knowledge retrieval.
| CRACK | Base | Delta | |
|---|---|---|---|
| Total | 41/65 (63.1%) | 37/65 (56.9%) | +6.2% |
Install & Usage
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-VL-4B-JANG_4S-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 Mode
Thinking is ON by default. To disable:
prompt = tokenizer.apply_chat_template(
messages, add_generation_prompt=True,
enable_thinking=False, tokenize=False)
Tip: Use
temperature=1.0for chat. Usetemperature=0.0for structured tasks like MMLU.
About JANG
JANG (Jang Adaptive N-bit Grading) is a mixed-precision quantization format for Apple Silicon — the GGUF equivalent for MLX.
About CRACK
CRACK (Controlled Refusal Ablation via Calibrated Knockouts) removes safety alignment from LLMs at the weight level using per-layer projected vectors from 512 structurally-mirrored prompt pairs.
Links
Disclaimer
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.
한국어
Qwen 3.5 VL 4B — JANG_4S + CRACK
| 항목 | 내용 |
|---|---|
| 크기 | 3 GB |
| HarmBench | 91.2% (292/320) |
| MMLU | 63.1% (기본 56.9% 대비 +6.2%) |
| 속도 | 134 tok/s (M4 Max) |
| 비전 | 지원 (MLX Studio / vMLX) |
| 최소 요구사양 | 8 GB 메모리 Mac |
pip install "jang[mlx]"
GitHub · HuggingFace · MLX Studio · Ko-fi · X @dealignai