Instructions to use benchang1110/Taiwan-tinyllama-v1.0-chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use benchang1110/Taiwan-tinyllama-v1.0-chat with Transformers:

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

pipe = pipeline("text-generation", model="benchang1110/Taiwan-tinyllama-v1.0-chat")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("benchang1110/Taiwan-tinyllama-v1.0-chat")
model = AutoModelForCausalLM.from_pretrained("benchang1110/Taiwan-tinyllama-v1.0-chat", device_map="auto")
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 Settings

vLLM

How to use benchang1110/Taiwan-tinyllama-v1.0-chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "benchang1110/Taiwan-tinyllama-v1.0-chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "benchang1110/Taiwan-tinyllama-v1.0-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/benchang1110/Taiwan-tinyllama-v1.0-chat

SGLang

How to use benchang1110/Taiwan-tinyllama-v1.0-chat 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 "benchang1110/Taiwan-tinyllama-v1.0-chat" \
    --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": "benchang1110/Taiwan-tinyllama-v1.0-chat",
		"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 "benchang1110/Taiwan-tinyllama-v1.0-chat" \
        --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": "benchang1110/Taiwan-tinyllama-v1.0-chat",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use benchang1110/Taiwan-tinyllama-v1.0-chat with Docker Model Runner:
```
docker model run hf.co/benchang1110/Taiwan-tinyllama-v1.0-chat
```

Model Card for Model ID

This model is the instruction finetuning version of benchang1110/Taiwan-tinyllama-v1.0-base.

Usage

import torch, transformers  

def generate_response():
    model = transformers.AutoModelForCausalLM.from_pretrained("benchang1110/Taiwan-tinyllama-v1.0-chat", torch_dtype=torch.bfloat16, device_map=device,attn_implementation="flash_attention_2")
    tokenizer = transformers.AutoTokenizer.from_pretrained("benchang1110/Taiwan-tinyllama-v1.0-chat")
    streamer = transformers.TextStreamer(tokenizer,skip_prompt=True)
    while(1):
        prompt = input('USER:')
        if prompt == "exit":
            break
        print("Assistant: ")
        message = [
           {'content': prompt, 'role': 'user'},
        ]
        untokenized_chat = tokenizer.apply_chat_template(message,tokenize=False,add_generation_prompt=False)
        inputs = tokenizer.encode_plus(untokenized_chat, add_special_tokens=True, return_tensors="pt",return_attention_mask=True).to(device)
        outputs = model.generate(inputs["input_ids"],attention_mask=inputs['attention_mask'],streamer=streamer,use_cache=True,max_new_tokens=512,do_sample=True,temperature=0.1,repetition_penalty=1.2)


if __name__ == '__main__':
    device = 'cuda' if torch.cuda.is_available() else 'cpu'
    generate_response()

Downloads last month: 5

Safetensors

Model size

1B params

Tensor type

BF16

benchang1110
/

Taiwan-tinyllama-v1.0-chat

Model Card for Model ID

Usage

Dataset used to train benchang1110/Taiwan-tinyllama-v1.0-chat