Instructions to use Hai929/The_GuageLLM_23M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Hai929/The_GuageLLM_23M with Transformers:

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

pipe = pipeline("text-generation", model="Hai929/The_GuageLLM_23M")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Hai929/The_GuageLLM_23M")
model = AutoModelForCausalLM.from_pretrained("Hai929/The_GuageLLM_23M")
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 Hai929/The_GuageLLM_23M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Hai929/The_GuageLLM_23M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Hai929/The_GuageLLM_23M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Hai929/The_GuageLLM_23M

SGLang

How to use Hai929/The_GuageLLM_23M 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 "Hai929/The_GuageLLM_23M" \
    --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": "Hai929/The_GuageLLM_23M",
		"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 "Hai929/The_GuageLLM_23M" \
        --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": "Hai929/The_GuageLLM_23M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Hai929/The_GuageLLM_23M with Docker Model Runner:
```
docker model run hf.co/Hai929/The_GuageLLM_23M
```

GuageLLM-12M

GuageLLM-23M is a lightweight GPT-style language model (~23 million parameters) trained from scratch for experimentation, learning, and fast local inference.

This model is designed to be simple, transparent, and easy to run on CPUs while still demonstrating real transformer behavior.

🔹 Model Details

Architecture: GPT-2 style (decoder-only transformer)
Parameters: ~23M
Context Length: 64 tokens
Vocabulary Size: Custom tokenizer
Training: From scratch
Framework: 🤗 Transformers (PyTorch)

🔹 Intended Use

GuageLLM-23M is intended for:

Learning how transformers work internally
Small-scale text generation experiments
CPU-friendly inference
Research, education, and tinkering

⚠️ This model is not intended for production or safety-critical applications.

🔹 Usage

Text Generation (Pipeline)

from transformers import pipeline

pipe = pipeline(
    "text-generation",
    model="Hai929/GuageLLM_23M",
    trust_remote_code=True
)

pipe("The cat")

Downloads last month: 1,133

Safetensors

Model size

23M params

Tensor type

F32

Hai929
/

The_GuageLLM_23M

GuageLLM-12M

🔹 Model Details

🔹 Intended Use

🔹 Usage

Text Generation (Pipeline)

Space using Hai929/The_GuageLLM_23M 1