Downloads · 30 days
12
7% of all-time downloads
EXAONE-3.5-7.8B-Instruct-Yaho-5bit
ChanLumerico/EXAONE-3.5-7.8B-Instruct-Yaho-5bit
EXAONE-3.5-7.8B-Instruct-Yaho-5bit is a text generation model from ChanLumerico. Use it when you need the model to write or continue text. It is set up for mlx. The card lists the license as other.
5-bit MLX quantization of ChanLumerico/EXAONE-3.5-7.8B-Instruct-Yaho — a Korean gyaru-persona ('갸루귀신') fine-tune of LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct (LoRA SFT + custom ORPO). Persona, behavior triggers, and the ba…
All-time downloads
168
Public
Parameters
7.8B
5.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors5.9 GB · 100%
Parameter types
How the weights are stored.
U327.8B · 100%
Base models
Model card
From the Hugging Face model README
EXAONE-3.5-7.8B-Instruct-Yaho-5bit 🎀
5-bit MLX quantization of ChanLumerico/EXAONE-3.5-7.8B-Instruct-Yaho
— a Korean gyaru-persona ('갸루귀신') fine-tune of LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct
(LoRA SFT + custom ORPO). Persona, behavior triggers, and the baked 갸루귀신 name anchor are verified to
survive 5-bit quantization. Non-commercial (EXAONE AI Model License 1.1-NC).
The persona anchor is baked into chat_template.jinja — no system prompt needed.
pip install mlx-lm==0.29.1 "transformers==4.57.6"
from mlx_lm import load, generate
model, tok = load("ChanLumerico/EXAONE-3.5-7.8B-Instruct-Yaho-5bit")
p = tok.apply_chat_template([{"role": "user", "content": "오늘 시험 망쳤어…"}],
add_generation_prompt=True, tokenize=False)
print(generate(model, tok, prompt=p, max_tokens=512))
See the full model card / eval / interpretability and the project repo. Base: EXAONE-3.5 (LG AI Research).
- Task
- Text Generation
- Library
- mlx
- Type
- exaone
- License
- other
- Languages
- ko
- Created
- Jun 10, 2026
- Updated
- Jun 10, 2026