Instructions to use mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed with Transformers:

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

pipe = pipeline("text-generation", model="mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed")
model = AutoModelForCausalLM.from_pretrained("mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed

SGLang

How to use mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed 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 "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed with Docker Model Runner:
```
docker model run hf.co/mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

https://huggingface.co/HuggingFaceM4/tiny-random-LlamaForCausalLM

from sparseml.transformers import SparseAutoModelForCausalLM, SparseAutoTokenizer, oneshot
from sparseml.modifiers import SparseGPTModifier

model_id = "HuggingFaceM4/tiny-random-LlamaForCausalLM"
compressed_model_id = "mgoin/tiny-random-LlamaForCausalLM-pruned95-compressed"

# Apply SparseGPT to the model
oneshot(
    model=model_id,
    dataset="open_platypus",
    recipe=SparseGPTModifier(sparsity=0.95),
    output_dir="temp-output",
)

model = SparseAutoModelForCausalLM.from_pretrained("temp-output", torch_dtype="auto")
tokenizer = SparseAutoTokenizer.from_pretrained(model_id)

model.save_pretrained(compressed_model_id.split("/")[-1], save_compressed=True)
tokenizer.save_pretrained(compressed_model_id.split("/")[-1])

# Upload the checkpoint to Hugging Face
from huggingface_hub import HfApi
HfApi().upload_folder(
    folder_path=compressed_model_id.split("/")[-1],
    repo_id=compressed_model_id,
)

Downloads last month: 2

Safetensors

Model size

1.22M params

Tensor type

I64

F16