You can deploy this model and start earning money today!

bearzi/
MiMo-V2-Flash-JANG_2L

Name: MiMo-V2-Flash-JANG_2L
Availability: InStock

XiaomiMiMo/MiMo-V2-Flash

finetune

apache-2.0

Views

No views yet

MiMo-V2-Flash-JANG_2L

JANG adaptive mixed-precision MLX quantization produced via vmlx / jang-tools.

Quantization: 3.11b avg, profile JANG_2L, method mse, calibration weights
Profile: JANG_2L
Format: JANG v2 MLX safetensors
Compatible with: vmlx, MLX Studio, oMLX (with JANG patch)

Usage

vmlx (recommended)

bash

pip install 'vmlx[jang]'
vmlx serve bearzi/MiMo-V2-Flash-JANG_2L

Python

python

from jang_tools.loader import load_jang_model
from mlx_lm import generate

model, tokenizer = load_jang_model("bearzi/MiMo-V2-Flash-JANG_2L")
messages = [{"role": "user", "content": "Hello"}]
prompt = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=False)
print(generate(model, tokenizer, prompt=prompt, max_tokens=512, verbose=True))

About JANG

JANG (Jang Adaptive N-bit Grading) assigns different bit widths to different layer types — attention layers get more bits, MLP/expert layers compress harder. This preserves model coherence at aggressive compression levels where uniform quantization breaks down.

See JANG documentation and scores at jangq.ai.

Comparative benchmarks and feedback welcome — please open a discussion.