Nemotron-Cascade-2-30B-A3B-JANG_2L

IMO Gold Medal reasoning in 10 GB. Nemotron-Cascade-2 achieves 88% MMLU with reasoning at just 10 GB — fits on 16 GB MacBooks. Hybrid Mamba-2 SSM + MoE + Attention. Only 6 KV cache attention layers = minimal memory at long context.

LM Studio, Ollama, oMLX do NOT support JANG format. Use MLX Studio or pip install "jang[mlx]>=2.1.5".

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

Key Features

• 88.0% MMLU (200 questions, reasoning mode) — IMO Gold Medal model in 10 GB
• 130 tok/s generation, 112 tok/s prefill
• 10.3 GB on disk, 10.3 GB GPU RAM (peak 11.1 GB)
• Reasoning mode: <think>...</think> step-by-step problem solving
• Tiny KV cache: only 6 attention layers, 0.2 GB at 32K context
• Hybrid architecture: Mamba-2 SSM + MoE (128 experts, top-6) + Attention

Results: JANG vs MLX (200-question MMLU)

Per-subject comparison. All models tested with and without reasoning.

Summary

JANG_2L is the only quantization that fits 16 GB Macs while delivering 88% MMLU with reasoning. JANG_4M beats MLX 4-bit (93.0% vs 92.5%) at the same 17 GB size.

Also see: JANG_4M (17 GB) — same size as MLX 4-bit, higher reasoning score.

Specs

Requirements

• Apple Silicon Mac with 16+ GB unified memory
• MLX Studio or pip install "jang[mlx]>=2.1.5"

Quick Start

pip install "jang[mlx]>=2.1.5"

from jang_tools.loader import load_jang_model
from mlx_lm import generate

model, tokenizer = load_jang_model("JANGQ-AI/Nemotron-Cascade-2-30B-A3B-JANG_2L")

# With reasoning (recommended)
messages = [{"role": "user", "content": "Solve: what is the integral of x^2 * e^x?"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False,
    add_generation_prompt=True, enable_thinking=True)
result = generate(model, tokenizer, prompt=prompt, max_tokens=2048)

# Without reasoning (faster)
prompt = tokenizer.apply_chat_template(messages, tokenize=False,
    add_generation_prompt=True, enable_thinking=False)
result = generate(model, tokenizer, prompt=prompt, max_tokens=100)

Technical Notes

• Mamba-2 SSM: Most layers use state-space models, enabling efficient long-context with minimal KV cache.
• Only 6 attention layers: KV cache is tiny (0.2 GB at 32K). Most models use 25-100% attention layers.
• nemotron_h architecture: Requires JANG loader for proper weight mapping. Standard mlx-lm has incomplete support.
• IMO Gold Medal: This model achieves competition-level mathematical reasoning at 30B scale.