Instructions to use kzhou35/coldstart with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kzhou35/coldstart with Transformers:

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

pipe = pipeline("text-generation", model="kzhou35/coldstart")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("kzhou35/coldstart")
model = AutoModelForCausalLM.from_pretrained("kzhou35/coldstart")
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 kzhou35/coldstart with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kzhou35/coldstart"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kzhou35/coldstart",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kzhou35/coldstart

SGLang

How to use kzhou35/coldstart 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 "kzhou35/coldstart" \
    --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": "kzhou35/coldstart",
		"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 "kzhou35/coldstart" \
        --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": "kzhou35/coldstart",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use kzhou35/coldstart with Docker Model Runner:
```
docker model run hf.co/kzhou35/coldstart
```

coldstart

File size: 1,625 Bytes

4fe36e5

{"current_steps": 3, "total_steps": 20, "loss": 1.0435, "lr": 4.72751631047092e-05, "epoch": 0.2874251497005988, "percentage": 15.0, "elapsed_time": "0:05:37", "remaining_time": "0:31:54", "throughput": 1984.49, "total_tokens": 670576}
{"current_steps": 6, "total_steps": 20, "loss": 0.6701, "lr": 3.969463130731183e-05, "epoch": 0.5748502994011976, "percentage": 30.0, "elapsed_time": "0:11:13", "remaining_time": "0:26:11", "throughput": 2014.35, "total_tokens": 1356912}
{"current_steps": 9, "total_steps": 20, "loss": 0.4917, "lr": 2.8910861626005776e-05, "epoch": 0.8622754491017964, "percentage": 45.0, "elapsed_time": "0:16:49", "remaining_time": "0:20:33", "throughput": 2021.8, "total_tokens": 2040576}
{"current_steps": 12, "total_steps": 20, "loss": 0.4176, "lr": 1.7274575140626318e-05, "epoch": 1.095808383233533, "percentage": 60.0, "elapsed_time": "0:21:22", "remaining_time": "0:14:15", "throughput": 2019.6, "total_tokens": 2590528}
{"current_steps": 15, "total_steps": 20, "loss": 0.3634, "lr": 7.3223304703363135e-06, "epoch": 1.3832335329341316, "percentage": 75.0, "elapsed_time": "0:26:59", "remaining_time": "0:08:59", "throughput": 2022.79, "total_tokens": 3275616}
{"current_steps": 18, "total_steps": 20, "loss": 0.3434, "lr": 1.2235870926211619e-06, "epoch": 1.6706586826347305, "percentage": 90.0, "elapsed_time": "0:32:34", "remaining_time": "0:03:37", "throughput": 2025.83, "total_tokens": 3959408}
{"current_steps": 20, "total_steps": 20, "epoch": 1.8622754491017965, "percentage": 100.0, "elapsed_time": "0:36:39", "remaining_time": "0:00:00", "throughput": 2008.18, "total_tokens": 4416112}