Instructions to use lianghsun/Marble-3B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lianghsun/Marble-3B-Instruct with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="lianghsun/Marble-3B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lianghsun/Marble-3B-Instruct")
model = AutoModelForCausalLM.from_pretrained("lianghsun/Marble-3B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lianghsun/Marble-3B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lianghsun/Marble-3B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lianghsun/Marble-3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lianghsun/Marble-3B-Instruct

SGLang

How to use lianghsun/Marble-3B-Instruct with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "lianghsun/Marble-3B-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lianghsun/Marble-3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "lianghsun/Marble-3B-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lianghsun/Marble-3B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lianghsun/Marble-3B-Instruct with Docker Model Runner:
```
docker model run hf.co/lianghsun/Marble-3B-Instruct
```

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Model Card for Marble-3B-Instruct

Marble-3B-Instruct 是 Marble-3B（Granite 3.1 3B-A800M MoE 之繁中 CPT 版本）的指令微調版本，以繁中對話資料完成 SFT，提供 MoE 架構下的台灣語境繁中對話能力。

⚠️ 規格重點：本模型為 3B Mixture-of-Experts（MoE）模型、純文本單模態。

Model Details

MoE 架構在推論時僅活化部分專家權重，可在保留模型容量的同時降低推論成本。本模型在 Marble-3B 之繁中 CPT 底座上做指令微調，目標是讓 MoE 架構在繁中與台灣語境下提供穩定可用的對話能力。

核心特點 (Key Features)

MoE 架構對話模型：推論成本相對於同容量稠密模型更低。
台灣語境對齊：訓練資料以繁中與台灣常見任務為主。
可端側部署：3B 規模友善於消費級硬體。

Model Description

Developed by: Liang Hsun Huang
Funded by: APMIC
Base model: lianghsun/Marble-3B
Model type: GraniteMoeForCausalLM (Transformers)
Language(s) (NLP): Traditional Chinese, English
License: Apache-2.0
Finetuned from model: lianghsun/Marble-3B

Model Sources

Repository: lianghsun/Marble-3B-Instruct

Citation

@misc{marble_3b_instruct,
  title        = {Marble-3B-Instruct: A Traditional Chinese Instruction-Tuned Granite 3B-A800M MoE Model for Taiwan},
  author       = {Huang, Liang Hsun},
  year         = {2025},
  howpublished = {\url{https://huggingface.co/lianghsun/Marble-3B-Instruct}}
}