gemma-4-26B-A4B-ko-sft-v2.1 – AI Model by jiwon9703 | AlphaNeural AI

You can deploy this model and start earning money today!

jiwon9703/
gemma-4-26B-A4B-ko-sft-v2.1

like

0

image-text-to-text

unsloth/gemma-4-26B-A4B-it

endpoints_compatible

Views

No views yet

gemma-4-26B-A4B-ko-sft-v2.1

Gemma4-26B-A4B 한국어 SFT v2.1 — v2 데이터(6,750) + 표기 정정 + LoRA r=64 α=64 (v2 r=16 α=32 대비 capacity 4x).

모델 정보

항목	내용
Base Model	unsloth/gemma-4-26B-A4B-it
학습 방법	LoRA SFT (Unsloth + TRL)
프레임워크	transformers, peft
라이센스	Apache 2.0

사용법

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("jiwon9703/gemma-4-26B-A4B-ko-sft-v2.1")
tokenizer = AutoTokenizer.from_pretrained("jiwon9703/gemma-4-26B-A4B-ko-sft-v2.1")

vLLM 서빙

vllm serve jiwon9703/gemma-4-26B-A4B-ko-sft-v2.1 --max-model-len 8192 --reasoning-parser gemma4