Instructions to use ONYULTEA/K-Decision-9B-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ONYULTEA/K-Decision-9B-v2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("zero-shot-classification", model="ONYULTEA/K-Decision-9B-v2")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("ONYULTEA/K-Decision-9B-v2") model = AutoModelForCausalLM.from_pretrained("ONYULTEA/K-Decision-9B-v2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
K-Decision-9B v2
Korean-specialized decision model by ONYULTEA
In collaboration with Hallym University
K-Decision is a lightweight decision engine that takes a state and typed questions as input and returns structured decisions with calibrated probability distributions over all options β all in a single forward pass. System One / Jev compatible.
Benchmarks
| Benchmark | Score |
|---|---|
| LocalLLaMA typed-decisions (2,000 decisions, fitted) | 0.8295 |
| KoBEST 5-task conversion (2,600 items) | 0.866 |
| Korean Life-Domain Holdout (7 domains) | 0.997 |
| korean-jev-decision-index (3,000 items) | 0.635 |
| Decision Index 0.3 public suite (140K requests) | Index 30.7 Β· 0 errors Β· p50 168ms |
- Trained and served on a single 24GB consumer GPU (int8 LoRA, r=32)
Quickstart
/v1/systemone server
git clone https://github.com/onyultea/k-decision
python src/serve_systemone.py --merged ONYULTEA/K-Decision-9B-v2 --port 8111
// Request
{"state": "Civil complaint: Strong gas smell (call center)",
"questions": {
"route": {"type": "choice", "instructions": "Which department?",
"criteria": {"gas": "Gas Safety", "water": "Waterworks", "road": "Roads"}},
"urgent": {"type": "noul", "instructions": "Urgent?"}}}
// Response β probability for every option
{"answers": {
"route": {"type": "choice", "choice": "gas",
"probabilities": {"gas": 0.87, "water": 0.05, "road": 0.08}},
"urgent": {"type": "noul", "noul": 0.93}}}
Acknowledgments
We sincerely thank Prof. Youngwoong Ko (School of Software, Hallym University) for his invaluable guidance and support throughout this fine-tuning research. This work was conducted as an industryβacademia collaboration between ONYULTEA and Hallym University. In addition, Hallym University regulations were used as reference material for the compliance-decision training and evaluation in the development and research of K-Decision.
K-Decision-9B v2 (νκ΅μ΄)
ONYULTEA νκ΅ νΉν λμμ λͺ¨λΈ
νλ¦Όλνκ΅ μ°ννλ ₯
μν(state)μ νμ΄νλ μ§λ¬Έμ μ λ ₯λ°μ λͺ¨λ μ΅μ μ 보μ λ νλ₯ λΆν¬λ₯Ό λ¨μΌ forward ν¨μ€λ‘ λ°ν (System One / Jev νΈν).
μ±λ₯
| λ²€μΉλ§ν¬ | μ μ |
|---|---|
| LocalLLaMA typed-decisions (2,000κ²°μ ) | 0.8295 |
| KoBEST 5νμ€ν¬ | 0.866 |
| νκ΅ μν λλ©μΈ (7λλ©μΈ) | 0.997 |
| korean-jev-decision-index | 0.635 |
| Decision Index 0.3 (14λ§ μμ²) | μ§μ 30.7 Β· μλ¬ 0 |
λΉ λ₯Έ μμ
git clone https://github.com/onyultea/k-decision
python src/serve_systemone.py --merged ONYULTEA/K-Decision-9B-v2 --port 8111
κ°μ¬μ κΈ
λ³Έ μ°κ΅¬μ νμΈνλ κ³Όμ μ λ°μ κ±Έμ³ κ·ν μ‘°μΈκ³Ό μ§μμ 보λ΄μ£Όμ **νλ¦Όλνκ΅ μννΈμ¨μ΄μ΅ν©λν κ³ μμ κ΅μλ**κ» μ§μ¬μΌλ‘ κ°μ¬λ립λλ€. λ³Έ μ±κ³Όλ (μ£Ό)μ¨μ¨ν°μ νλ¦Όλνκ΅μ μ°ννλ ₯μΌλ‘ μ§νλμμ΅λλ€. λν λ³Έ K-Decisionμ κ°λ°Β·μ°κ΅¬μλ νλ¦Όλνκ΅ κ·μ μ΄ κ·μ μ€κ±° νλ¨(regulation compliance decision) νμ΅Β·νκ°μ μ°Έμ‘° μλ£λ‘ νμ©λμμ΅λλ€.
Β© 2026 ONYULTEA Γ Hallym University Β· https://onyultea.kr
- Downloads last month
- 73
Model tree for ONYULTEA/K-Decision-9B-v2
Space using ONYULTEA/K-Decision-9B-v2 1
Evaluation results
- accuracy (fitted) on LocalLLaMA/typed-decisions (test, 400 cases / 2,000 decisions)self-reported0.830