Instructions to use MBZUAI/MobiLlama-1B-Chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MBZUAI/MobiLlama-1B-Chat with Transformers:

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

pipe = pipeline("text-generation", model="MBZUAI/MobiLlama-1B-Chat", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MBZUAI/MobiLlama-1B-Chat", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("MBZUAI/MobiLlama-1B-Chat", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MBZUAI/MobiLlama-1B-Chat with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MBZUAI/MobiLlama-1B-Chat"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MBZUAI/MobiLlama-1B-Chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/MBZUAI/MobiLlama-1B-Chat

SGLang

How to use MBZUAI/MobiLlama-1B-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 "MBZUAI/MobiLlama-1B-Chat" \
    --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": "MBZUAI/MobiLlama-1B-Chat",
		"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 "MBZUAI/MobiLlama-1B-Chat" \
        --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": "MBZUAI/MobiLlama-1B-Chat",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use MBZUAI/MobiLlama-1B-Chat with Docker Model Runner:
```
docker model run hf.co/MBZUAI/MobiLlama-1B-Chat
```

MLP weights in each layer seem to be different

by Jeethu - opened May 16, 2024

Discussion

Jeethu

May 16, 2024

Thanks for releasing the MobiLlama models! IIUC, the paper states that the FFN blocks are shared across all transformer blocks. I've verified that this in indeed the case with the 0.5B model and the MLP weights are identical in each layer, but it doesn't seem to be the case with this model. Any reason for this discrepancy?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment