Instructions to use abacusai/Smaug-72B-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use abacusai/Smaug-72B-v0.1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="abacusai/Smaug-72B-v0.1")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("abacusai/Smaug-72B-v0.1") model = AutoModelForCausalLM.from_pretrained("abacusai/Smaug-72B-v0.1") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use abacusai/Smaug-72B-v0.1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "abacusai/Smaug-72B-v0.1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "abacusai/Smaug-72B-v0.1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/abacusai/Smaug-72B-v0.1
- SGLang
How to use abacusai/Smaug-72B-v0.1 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 "abacusai/Smaug-72B-v0.1" \ --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": "abacusai/Smaug-72B-v0.1", "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 "abacusai/Smaug-72B-v0.1" \ --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": "abacusai/Smaug-72B-v0.1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use abacusai/Smaug-72B-v0.1 with Docker Model Runner:
docker model run hf.co/abacusai/Smaug-72B-v0.1
Smaug - Japanese Language Support
Hi team,
I was wondering if Smaug model included Japanese datasets during its training phase. If Yes, could you please the Japanese contents on which Smaug model has been trained?
We did not utilise any Japanese datasets during the training of Smaug, and it does not appear as though the model we started from (https://huggingface.co/moreh/MoMo-72B-lora-1.8.7-DPO) did either.
Once we release our technique paper in a couple of weeks though you could try to replicate the process with some Japanese datasets added in :)
Yes no decent Japanese open source LLMs exist that would be nice