Instructions to use bigscience/bloom-1b1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use bigscience/bloom-1b1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="bigscience/bloom-1b1")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("bigscience/bloom-1b1") model = AutoModelForCausalLM.from_pretrained("bigscience/bloom-1b1") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use bigscience/bloom-1b1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "bigscience/bloom-1b1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "bigscience/bloom-1b1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/bigscience/bloom-1b1
- SGLang
How to use bigscience/bloom-1b1 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 "bigscience/bloom-1b1" \ --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": "bigscience/bloom-1b1", "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 "bigscience/bloom-1b1" \ --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": "bigscience/bloom-1b1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use bigscience/bloom-1b1 with Docker Model Runner:
docker model run hf.co/bigscience/bloom-1b1
System Requirements
#38
by Robo0890 - opened
I’m hoping to fine tune this model on custom data for an upcoming project. I have an RTX-3060 and 16GB of RAM, should I be able to use this model for training or should I go with the 560m parameter version?
In my experience, fine-tuning this model need 32GB GPU RAM minimum. I fine-tuned this model on Google Colab Pro+. This task expense a lot of money.
Yeah, I’ll just stick with the little one
I recommend using the bloom 560m. It works fine as the same 1b1 one but need less GPU RAM.