Instructions to use facebook/MobileLLM-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/MobileLLM-1B with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="facebook/MobileLLM-1B", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("facebook/MobileLLM-1B", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use facebook/MobileLLM-1B with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "facebook/MobileLLM-1B" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "facebook/MobileLLM-1B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/facebook/MobileLLM-1B
- SGLang
How to use facebook/MobileLLM-1B 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 "facebook/MobileLLM-1B" \ --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": "facebook/MobileLLM-1B", "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 "facebook/MobileLLM-1B" \ --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": "facebook/MobileLLM-1B", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use facebook/MobileLLM-1B with Docker Model Runner:
docker model run hf.co/facebook/MobileLLM-1B
a horrible function in `modeling_mobilellm.py`
#5
by J22 - opened
if not (...., False) is killing me.
def get_output_embeddings(self):
return (
self.lm_head
if not getattr(self.config, "share_embedding", False)
else self.get_input_embeddings()
)
I would do it this way:
def get_output_embeddings(self):
return (
self.get_input_embeddings()
if getattr(self.config, "share_embedding", False)
else self.lm_head
)
This is even better, but default value is changed:
def get_output_embeddings(self):
return (
self.get_input_embeddings()
if getattr(self.config, "share_embedding", True)
else self.lm_head
)
Thanks for pointing that out. 😆 We modified modeling.py!
zechunliu changed discussion status to closed