JANGQ-AI/Qwen3.5-122B-A10B-JANG_2S

CRITICAL FIX (2026-03-19): Fixed eos_token_id — previous versions caused infinite thinking loops. You MUST re-download this model if you downloaded before today.

Update (2026-03-18): Models have been updated to v2.1.0 with VLM support, proper tokenizer, and fixed configs. If you downloaded before this date, please re-download for full MLX Studio compatibility.

MLX Studio — the only app that natively supports JANG models

---

Early Adoption: LM Studio, Ollama, oMLX, Inferencer do not support JANG yet. Use MLX Studio or pip install "jang[mlx]". Ask your favorite app's creators to add JANG support!

---

Qwen3.5-122B-A10B — JANG_2S (MoE, 2-bit) — VLM

JANG — Jang Adaptive N-bit Grading | Mixed-Precision Quantization for Apple Silicon

     

JANG is fully open-source. Quantization engine, research, and full commit history: github.com/jjang-ai/jangq. Created by Jinho Jang.

Results (200-question MMLU)

Model	MMLU	Size
JANG_4K	86%	69 GB
JANG_2S	79%	38 GB
MLX 4-bit	85%	64 GB
MLX 2-bit	56.5%	36 GB

JANG_2S at 38 GB scores 79% while MLX 2-bit at 36 GB scores 56.5%. +22.5 points at nearly the same size. On MoE models with 256 experts, JANG's tier-based allocation protects the <2% of critical parameters while compressing the 98% expert MLP to 2-bit.

Specs

Metric	Value
Source	Qwen3.5-122B-A10B
Architecture	MoE (256 experts, 8 active) + GatedDeltaNet SSM
Profile	JANG_2S (CRITICAL=6, IMPORTANT=4, COMPRESS=2)
GPU Memory	~35 GB
Best for	64+ GB Mac
VLM	Yes
Speed	54 tok/s
Format	v2 (MLX-native, instant load)

Install

pip install "jang[mlx]"

For Vision-Language models:

pip install "jang[vlm]"

Quick Start

from jang_tools.loader import load_jang_model
from mlx_lm.sample_utils import make_sampler
from mlx_lm.generate import generate_step
import mlx.core as mx

model, tokenizer = load_jang_model("JANGQ-AI/Qwen3.5-122B-A10B-JANG_2S")
sampler = make_sampler(temp=0.7)

tokens = tokenizer.encode("What is photosynthesis?")
for tok, _ in generate_step(prompt=mx.array(tokens), model=model, max_tokens=200, sampler=sampler):
    t = tok.item() if hasattr(tok, 'item') else int(tok)
    print(tokenizer.decode([t]), end="", flush=True)
    if t == tokenizer.eos_token_id:
        break

VLM Inference

from jang_tools.loader import load_jang_vlm_model
from mlx_vlm import generate

model, processor = load_jang_vlm_model("JANGQ-AI/Qwen3.5-122B-A10B-JANG_2S")

prompt = processor.tokenizer.apply_chat_template(
    [{"role": "user", "content": [
        {"type": "image", "image": "photo.jpg"},
        {"type": "text", "text": "Describe this image."}
    ]}], add_generation_prompt=True, tokenize=False, enable_thinking=False)

result = generate(model, processor, prompt, ["photo.jpg"], max_tokens=200)
print(result.text)

Links

• GitHub | HuggingFace | MLX Studio | PyPI | Format Spec

---

---

한국어

Qwen3.5-122B (MoE) — JANG 2S

JANG은 Apple Silicon을 위한 혼합정밀도 양자화 포맷입니다. MLX를 위한 GGUF와 같은 역할을 합니다.

모델	MMLU	크기
JANG_2S	79%	38 GB
MLX 2-bit	56.5%	36 GB

설치

pip install "jang[mlx]"

호환성

현재 **MLX Studio**만 JANG 포맷을 기본 지원합니다. LM Studio, Ollama 등은 아직 지원하지 않습니다.

GitHub · HuggingFace · MLX Studio · PyPI

---

장진호 제작 · Created by Jinho Jang — jangq.ai · @dealignai